Publish fully validated per-task Croissant metadata and online validation reports

croissant/SimplexTasks-12_Croissant_VALIDATED.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3cb748ad674a96b1da362ea1ad864c0d91f591246a1aa84eca1933c60125dec9
+size 46989

croissant/affectivetext_emotions.json

@@ -0,0 +1,362 @@
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/affectivetext_emotions/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#affectivetext_emotions",
+  "name": "SimplexTasks-12 — SemEval AffectiveText emotions",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/derived/affectivetext",
+  "description": "SemEval AffectiveText emotions is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 1000 samples with 6 simplex components. Frozen predictor: Frozen zero-shot emotion scorer. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/affectivetext.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "affectivetext_emotions"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "affectivetext_emotions.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/affectivetext_emotions.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "35526 B",
+      "sha256": "14285f9b95857ff2bb91b005df1d3763084a2b3b91408bc6bdf92e577882d7b0"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "1000 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_3",
+          "name": "U_3",
+          "description": "Frozen predictor output, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_4",
+          "name": "U_4",
+          "description": "Frozen predictor output, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_5",
+          "name": "U_5",
+          "description": "Frozen predictor output, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_3",
+          "name": "Y_3",
+          "description": "Benchmark target composition, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_4",
+          "name": "Y_4",
+          "description": "Benchmark target composition, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_5",
+          "name": "Y_5",
+          "description": "Benchmark target composition, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_5"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "The reported benchmark tables use a frozen closed-model cache; the release omits raw headlines and raw API responses. A fully open TF-IDF+SVD+kNN fallback cache-builder is included for local reproduction, but it is not identical to the main predictor used in the paper tables. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "English headline selection, subjective emotion labels, and the frozen model may introduce language, cultural and annotation biases. This task does not measure an individual’s emotional state.",
+  "rai:personalSensitiveInformation": "Derived emotion mixtures and headline identifiers are released; raw headlines and raw API responses are excluded. Emotion scores describe headlines, not diagnosed mental states.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper’s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": false,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://aclanthology.org/S07-1013/"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/affectivetext/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/affectivetext/Y.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-affectivetext_emotions",
+    "description": "Normalized six-emotion gold scores from SemEval-2007 AffectiveText paired with a frozen zero-shot scorer cache. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. The package omits raw headlines and raw API responses, and keeps only derived score arrays. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/affectivetext.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Normalized six-emotion gold scores from SemEval-2007 AffectiveText paired with a frozen zero-shot scorer cache. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. The package omits raw headlines and raw API responses, and keeps only derived score arrays. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/affectivetext.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}

croissant/cifar10_softmax.json

@@ -0,0 +1,482 @@
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/cifar10_softmax/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#cifar10_softmax",
+  "name": "SimplexTasks-12 — CIFAR-10 softmax",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/derived/cifar10_softmax",
+  "description": "CIFAR-10 softmax is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 10000 samples with 10 simplex components. Frozen predictor: Frozen ResNet-18 softmax cache. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/cifar10_softmax.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "cifar10_softmax"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "cifar10_softmax.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/cifar10_softmax.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "590269 B",
+      "sha256": "23d39b4db178de94ee8b76d8c64e0817ca0c9404a5bfec98369b68320b0ef5dc"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "10000 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_3",
+          "name": "U_3",
+          "description": "Frozen predictor output, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_4",
+          "name": "U_4",
+          "description": "Frozen predictor output, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_5",
+          "name": "U_5",
+          "description": "Frozen predictor output, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_6",
+          "name": "U_6",
+          "description": "Frozen predictor output, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_7",
+          "name": "U_7",
+          "description": "Frozen predictor output, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_8",
+          "name": "U_8",
+          "description": "Frozen predictor output, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_9",
+          "name": "U_9",
+          "description": "Frozen predictor output, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_9"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_3",
+          "name": "Y_3",
+          "description": "Benchmark target composition, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_4",
+          "name": "Y_4",
+          "description": "Benchmark target composition, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_5",
+          "name": "Y_5",
+          "description": "Benchmark target composition, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_6",
+          "name": "Y_6",
+          "description": "Benchmark target composition, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_7",
+          "name": "Y_7",
+          "description": "Benchmark target composition, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_8",
+          "name": "Y_8",
+          "description": "Benchmark target composition, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_9",
+          "name": "Y_9",
+          "description": "Benchmark target composition, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_9"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "This is a classification proxy rather than a naturally continuous composition task. The release ships derived arrays only and does not mirror raw CIFAR-10 images. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "The ten selected image classes, image collection and frozen ResNet-18 may introduce selection and model biases; this is not representative of all visual recognition settings.",
+  "rai:personalSensitiveInformation": "Released U and Y contain class probabilities and one-hot targets; raw images are excluded. No direct personal identifiers are described in these arrays.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper’s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": false,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://www.cs.toronto.edu/~kriz/cifar.html"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/cifar10_softmax/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/cifar10_softmax/Y.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-cifar10_softmax",
+    "description": "A 10-way one-hot class distribution paired with a frozen ResNet-18 softmax cache. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. No raw images are redistributed in this package. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/cifar10_softmax.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "A 10-way one-hot class distribution paired with a frozen ResNet-18 softmax cache. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. No raw images are redistributed in this package. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/cifar10_softmax.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}

croissant/d1_homogeneous.json

@@ -0,0 +1,308 @@
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d1_homogeneous/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#d1_homogeneous",
+  "name": "SimplexTasks-12 — d1 homogeneous",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/synthetic/D1",
+  "description": "d1 homogeneous is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 7500 samples with 3 simplex components. Synthetic regime: Homogeneous negative control. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/synthetic/D1_homogeneous.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "d1_homogeneous"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "d1_homogeneous.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/d1_homogeneous.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "295831 B",
+      "sha256": "9fed050352b82559a340d492cb51dbf3f554f15cc0724b004df45319dee38d49"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "7500 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/R",
+          "name": "R",
+          "description": "Frozen residual score.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "R"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sigma_true",
+          "name": "sigma_true",
+          "description": "True local scale from the synthetic generator.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sigma_true"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "Intended to confirm that the benchmark does not manufacture allocation failures when the DGP is homogeneous. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "Synthetic diagnostic design deliberately selects one heterogeneity pattern; it does not sample a human population or establish demographic representativeness.",
+  "rai:personalSensitiveInformation": "Generated numerical simulations; no human-subject records or personal attributes are used in these synthetic arrays.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper’s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": true,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D1.yaml"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D1/R.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D1/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D1/Y.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D1/sigma_true.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-d1_homogeneous",
+    "description": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D1.yaml. Configuration:\nexperiment: d1_homogeneous\n\ndgp:\n  name: pure_scale\n  K: 3\n  sigma_min: 0.2\n  c: 0.0\n  d_x: 2\n\ndata:\n  n_train: 1000\n  n_cal: 1000\n  n_scale_est: 500\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - partition\n  - twostage\n  - oneshot\n  - trainres\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D1.yaml. Configuration:\nexperiment: d1_homogeneous\n\ndgp:\n  name: pure_scale\n  K: 3\n  sigma_min: 0.2\n  c: 0.0\n  d_x: 2\n\ndata:\n  n_train: 1000\n  n_cal: 1000\n  n_scale_est: 500\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - partition\n  - twostage\n  - oneshot\n  - trainres\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}

croissant/d2_pure_scale.json

@@ -0,0 +1,308 @@
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d2_pure_scale/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#d2_pure_scale",
+  "name": "SimplexTasks-12 — d2 pure scale",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/synthetic/D2",
+  "description": "d2 pure scale is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 6250 samples with 3 simplex components. Synthetic regime: Smooth scale heterogeneity. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/synthetic/D2_smooth_scale.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "d2_pure_scale"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "d2_pure_scale.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/d2_pure_scale.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "278811 B",
+      "sha256": "de27d8fa6b0841f45fc9336fa717423d36e5f605145bc7001ee78e8076b52c2a"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "6250 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/R",
+          "name": "R",
+          "description": "Frozen residual score.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "R"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sigma_true",
+          "name": "sigma_true",
+          "description": "True local scale from the synthetic generator.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sigma_true"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "This regime isolates scale variation only; it is not intended to cover bias-type misspecification or changing tail shape. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "Synthetic diagnostic design deliberately selects one heterogeneity pattern; it does not sample a human population or establish demographic representativeness.",
+  "rai:personalSensitiveInformation": "Generated numerical simulations; no human-subject records or personal attributes are used in these synthetic arrays.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper’s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": true,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D2.yaml"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D2/R.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D2/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D2/Y.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D2/sigma_true.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-d2_pure_scale",
+    "description": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D2.yaml. Configuration:\nexperiment: d2_pure_scale\n\ndgp:\n  name: pure_scale\n  K: 3\n  sigma_min: 0.1\n  c: 0.5\n  d_x: 2\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - fullcp\n  - jackknife_plus\n  - partition\n  - twostage\n  - oneshot\n  - trainres\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D2.yaml. Configuration:\nexperiment: d2_pure_scale\n\ndgp:\n  name: pure_scale\n  K: 3\n  sigma_min: 0.1\n  c: 0.5\n  d_x: 2\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - fullcp\n  - jackknife_plus\n  - partition\n  - twostage\n  - oneshot\n  - trainres\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}

croissant/d3_discrete_groups_aligned.json

@@ -0,0 +1,518 @@
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d3_discrete_groups_aligned/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#d3_discrete_groups_aligned",
+  "name": "SimplexTasks-12 — d3 discrete groups aligned",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/synthetic/D3",
+  "description": "d3 discrete groups aligned is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 6250 samples with 10 simplex components. Synthetic regime: Aligned discrete groups. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/synthetic/D3_aligned_discrete.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "d3_discrete_groups_aligned"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "d3_discrete_groups_aligned.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/d3_discrete_groups_aligned.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "715419 B",
+      "sha256": "f6daa6cfb82fb63b206efb5d8211d200df9dc46552a9656d02d9fa7942344298"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "6250 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/R",
+          "name": "R",
+          "description": "Frozen residual score.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "R"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_3",
+          "name": "U_3",
+          "description": "Frozen predictor output, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_4",
+          "name": "U_4",
+          "description": "Frozen predictor output, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_5",
+          "name": "U_5",
+          "description": "Frozen predictor output, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_6",
+          "name": "U_6",
+          "description": "Frozen predictor output, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_7",
+          "name": "U_7",
+          "description": "Frozen predictor output, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_8",
+          "name": "U_8",
+          "description": "Frozen predictor output, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_9",
+          "name": "U_9",
+          "description": "Frozen predictor output, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_9"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_3",
+          "name": "Y_3",
+          "description": "Benchmark target composition, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_4",
+          "name": "Y_4",
+          "description": "Benchmark target composition, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_5",
+          "name": "Y_5",
+          "description": "Benchmark target composition, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_6",
+          "name": "Y_6",
+          "description": "Benchmark target composition, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_7",
+          "name": "Y_7",
+          "description": "Benchmark target composition, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_8",
+          "name": "Y_8",
+          "description": "Benchmark target composition, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_9",
+          "name": "Y_9",
+          "description": "Benchmark target composition, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_9"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sigma_true",
+          "name": "sigma_true",
+          "description": "True local scale from the synthetic generator.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sigma_true"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "The grouping is built into the DGP. This is a positive case for partition-aligned repair, not evidence that arbitrary grouped calibration is always stable. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "Synthetic diagnostic design deliberately selects one heterogeneity pattern; it does not sample a human population or establish demographic representativeness.",
+  "rai:personalSensitiveInformation": "Generated numerical simulations; no human-subject records or personal attributes are used in these synthetic arrays.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper’s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": true,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D3.yaml"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D3/R.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D3/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D3/Y.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D3/sigma_true.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-d3_discrete_groups_aligned",
+    "description": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D3.yaml. Configuration:\nexperiment: d3_discrete_groups_aligned\n\ndgp:\n  name: discrete_groups\n  K: 10\n  sigma_low: 0.08\n  sigma_high: 0.30\n  d_x: 5\n  easy_classes: 5\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - partition\n  - twostage\n  - fullcp\n  - jackknife_plus\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: argmax_group\n  n_strata: 2\n  split_index: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D3.yaml. Configuration:\nexperiment: d3_discrete_groups_aligned\n\ndgp:\n  name: discrete_groups\n  K: 10\n  sigma_low: 0.08\n  sigma_high: 0.30\n  d_x: 5\n  easy_classes: 5\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - partition\n  - twostage\n  - fullcp\n  - jackknife_plus\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: argmax_group\n  n_strata: 2\n  split_index: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}

croissant/d4_model_bias.json

@@ -0,0 +1,308 @@
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d4_model_bias/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#d4_model_bias",
+  "name": "SimplexTasks-12 — d4 model bias",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/synthetic/D4",
+  "description": "d4 model bias is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 6250 samples with 3 simplex components. Synthetic regime: Bias-type heterogeneity. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/synthetic/D4_bias_type.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "d4_model_bias"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "d4_model_bias.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/d4_model_bias.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "278534 B",
+      "sha256": "dfab38720fb1903807b2c2e90008d8b672ffe682a9dd184f7383acd7cccfba0f"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "6250 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/R",
+          "name": "R",
+          "description": "Frozen residual score.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "R"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sigma_true",
+          "name": "sigma_true",
+          "description": "True local scale from the synthetic generator.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sigma_true"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "This regime is not reducible to a single local scale field, so normalization cannot be expected to fully repair it. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "Synthetic diagnostic design deliberately selects one heterogeneity pattern; it does not sample a human population or establish demographic representativeness.",
+  "rai:personalSensitiveInformation": "Generated numerical simulations; no human-subject records or personal attributes are used in these synthetic arrays.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper’s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": true,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D4.yaml"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D4/R.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D4/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D4/Y.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D4/sigma_true.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-d4_model_bias",
+    "description": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D4.yaml. Configuration:\nexperiment: d4_model_bias\n\ndgp:\n  name: model_bias\n  K: 3\n  sigma_min: 0.1\n  c: 0.15\n  d_x: 2\n  bias_scale: 0.45\n  bias_type: rotational\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - fullcp\n  - jackknife_plus\n  - partition\n  - twostage\n  - oneshot\n  - trainres\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D4.yaml. Configuration:\nexperiment: d4_model_bias\n\ndgp:\n  name: model_bias\n  K: 3\n  sigma_min: 0.1\n  c: 0.15\n  d_x: 2\n  bias_scale: 0.45\n  bias_type: rotational\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - fullcp\n  - jackknife_plus\n  - partition\n  - twostage\n  - oneshot\n  - trainres\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}

croissant/d5_heavy_tail.json

@@ -0,0 +1,308 @@
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d5_heavy_tail/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#d5_heavy_tail",
+  "name": "SimplexTasks-12 — d5 heavy tail",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/synthetic/D5",
+  "description": "d5 heavy tail is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 6250 samples with 3 simplex components. Synthetic regime: Heavy-tail robustness. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/synthetic/D5_heavy_tail.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "d5_heavy_tail"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "d5_heavy_tail.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/d5_heavy_tail.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "278749 B",
+      "sha256": "dd7237032d30971d9097851d8b0484935618bb42c5eb84ed38e0f3e50d345a12"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "6250 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/R",
+          "name": "R",
+          "description": "Frozen residual score.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "R"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sigma_true",
+          "name": "sigma_true",
+          "description": "True local scale from the synthetic generator.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sigma_true"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "The regime is still synthetic and targeted; it does not span every possible deviation from the baseline score distribution. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "Synthetic diagnostic design deliberately selects one heterogeneity pattern; it does not sample a human population or establish demographic representativeness.",
+  "rai:personalSensitiveInformation": "Generated numerical simulations; no human-subject records or personal attributes are used in these synthetic arrays.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper’s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": true,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D5.yaml"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D5/R.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D5/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D5/Y.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D5/sigma_true.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-d5_heavy_tail",
+    "description": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D5.yaml. Configuration:\nexperiment: d5_heavy_tail\n\ndgp:\n  name: heavy_tail\n  K: 3\n  sigma_min: 0.1\n  c: 0.5\n  d_x: 2\n  df: 3.0\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - fullcp\n  - jackknife_plus\n  - partition\n  - twostage\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D5.yaml. Configuration:\nexperiment: d5_heavy_tail\n\ndgp:\n  name: heavy_tail\n  K: 3\n  sigma_min: 0.1\n  c: 0.5\n  d_x: 2\n  df: 3.0\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - fullcp\n  - jackknife_plus\n  - partition\n  - twostage\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}

croissant/d6_high_k.json

@@ -0,0 +1,1718 @@
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d6_high_k/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#d6_high_k",
+  "name": "SimplexTasks-12 — d6 high k",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/synthetic/D6",
+  "description": "d6 high k is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 17500 samples with 50 simplex components. Synthetic regime: High-dimensional simplex. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/synthetic/D6_highK.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "d6_high_k"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "d6_high_k.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/d6_high_k.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "10028544 B",
+      "sha256": "2e4fe45eae47a67c9d3023d8b302dcc0f0b9b8c7e007f3ab07e2db6011e0bbfc"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "17500 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/R",
+          "name": "R",
+          "description": "Frozen residual score.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "R"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_3",
+          "name": "U_3",
+          "description": "Frozen predictor output, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_4",
+          "name": "U_4",
+          "description": "Frozen predictor output, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_5",
+          "name": "U_5",
+          "description": "Frozen predictor output, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_6",
+          "name": "U_6",
+          "description": "Frozen predictor output, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_7",
+          "name": "U_7",
+          "description": "Frozen predictor output, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_8",
+          "name": "U_8",
+          "description": "Frozen predictor output, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_9",
+          "name": "U_9",
+          "description": "Frozen predictor output, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_9"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_10",
+          "name": "U_10",
+          "description": "Frozen predictor output, component 10 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_10"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_11",
+          "name": "U_11",
+          "description": "Frozen predictor output, component 11 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_11"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_12",
+          "name": "U_12",
+          "description": "Frozen predictor output, component 12 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_12"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_13",
+          "name": "U_13",
+          "description": "Frozen predictor output, component 13 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_13"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_14",
+          "name": "U_14",
+          "description": "Frozen predictor output, component 14 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_14"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_15",
+          "name": "U_15",
+          "description": "Frozen predictor output, component 15 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_15"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_16",
+          "name": "U_16",
+          "description": "Frozen predictor output, component 16 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_16"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_17",
+          "name": "U_17",
+          "description": "Frozen predictor output, component 17 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_17"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_18",
+          "name": "U_18",
+          "description": "Frozen predictor output, component 18 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_18"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_19",
+          "name": "U_19",
+          "description": "Frozen predictor output, component 19 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_19"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_20",
+          "name": "U_20",
+          "description": "Frozen predictor output, component 20 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_20"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_21",
+          "name": "U_21",
+          "description": "Frozen predictor output, component 21 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_21"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_22",
+          "name": "U_22",
+          "description": "Frozen predictor output, component 22 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_22"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_23",
+          "name": "U_23",
+          "description": "Frozen predictor output, component 23 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_23"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_24",
+          "name": "U_24",
+          "description": "Frozen predictor output, component 24 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_24"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_25",
+          "name": "U_25",
+          "description": "Frozen predictor output, component 25 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_25"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_26",
+          "name": "U_26",
+          "description": "Frozen predictor output, component 26 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_26"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_27",
+          "name": "U_27",
+          "description": "Frozen predictor output, component 27 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_27"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_28",
+          "name": "U_28",
+          "description": "Frozen predictor output, component 28 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_28"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_29",
+          "name": "U_29",
+          "description": "Frozen predictor output, component 29 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_29"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_30",
+          "name": "U_30",
+          "description": "Frozen predictor output, component 30 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_30"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_31",
+          "name": "U_31",
+          "description": "Frozen predictor output, component 31 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_31"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_32",
+          "name": "U_32",
+          "description": "Frozen predictor output, component 32 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_32"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_33",
+          "name": "U_33",
+          "description": "Frozen predictor output, component 33 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_33"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_34",
+          "name": "U_34",
+          "description": "Frozen predictor output, component 34 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_34"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_35",
+          "name": "U_35",
+          "description": "Frozen predictor output, component 35 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_35"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_36",
+          "name": "U_36",
+          "description": "Frozen predictor output, component 36 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_36"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_37",
+          "name": "U_37",
+          "description": "Frozen predictor output, component 37 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_37"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_38",
+          "name": "U_38",
+          "description": "Frozen predictor output, component 38 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_38"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_39",
+          "name": "U_39",
+          "description": "Frozen predictor output, component 39 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_39"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_40",
+          "name": "U_40",
+          "description": "Frozen predictor output, component 40 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_40"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_41",
+          "name": "U_41",
+          "description": "Frozen predictor output, component 41 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_41"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_42",
+          "name": "U_42",
+          "description": "Frozen predictor output, component 42 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_42"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_43",
+          "name": "U_43",
+          "description": "Frozen predictor output, component 43 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_43"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_44",
+          "name": "U_44",
+          "description": "Frozen predictor output, component 44 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_44"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_45",
+          "name": "U_45",
+          "description": "Frozen predictor output, component 45 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_45"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_46",
+          "name": "U_46",
+          "description": "Frozen predictor output, component 46 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_46"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_47",
+          "name": "U_47",
+          "description": "Frozen predictor output, component 47 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_47"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_48",
+          "name": "U_48",
+          "description": "Frozen predictor output, component 48 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_48"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_49",
+          "name": "U_49",
+          "description": "Frozen predictor output, component 49 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_49"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_3",
+          "name": "Y_3",
+          "description": "Benchmark target composition, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_4",
+          "name": "Y_4",
+          "description": "Benchmark target composition, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_5",
+          "name": "Y_5",
+          "description": "Benchmark target composition, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_6",
+          "name": "Y_6",
+          "description": "Benchmark target composition, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_7",
+          "name": "Y_7",
+          "description": "Benchmark target composition, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_8",
+          "name": "Y_8",
+          "description": "Benchmark target composition, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_9",
+          "name": "Y_9",
+          "description": "Benchmark target composition, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_9"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_10",
+          "name": "Y_10",
+          "description": "Benchmark target composition, component 10 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_10"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_11",
+          "name": "Y_11",
+          "description": "Benchmark target composition, component 11 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_11"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_12",
+          "name": "Y_12",
+          "description": "Benchmark target composition, component 12 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_12"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_13",
+          "name": "Y_13",
+          "description": "Benchmark target composition, component 13 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_13"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_14",
+          "name": "Y_14",
+          "description": "Benchmark target composition, component 14 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_14"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_15",
+          "name": "Y_15",
+          "description": "Benchmark target composition, component 15 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_15"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_16",
+          "name": "Y_16",
+          "description": "Benchmark target composition, component 16 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_16"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_17",
+          "name": "Y_17",
+          "description": "Benchmark target composition, component 17 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_17"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_18",
+          "name": "Y_18",
+          "description": "Benchmark target composition, component 18 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_18"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_19",
+          "name": "Y_19",
+          "description": "Benchmark target composition, component 19 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_19"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_20",
+          "name": "Y_20",
+          "description": "Benchmark target composition, component 20 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_20"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_21",
+          "name": "Y_21",
+          "description": "Benchmark target composition, component 21 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_21"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_22",
+          "name": "Y_22",
+          "description": "Benchmark target composition, component 22 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_22"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_23",
+          "name": "Y_23",
+          "description": "Benchmark target composition, component 23 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_23"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_24",
+          "name": "Y_24",
+          "description": "Benchmark target composition, component 24 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_24"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_25",
+          "name": "Y_25",
+          "description": "Benchmark target composition, component 25 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_25"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_26",
+          "name": "Y_26",
+          "description": "Benchmark target composition, component 26 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_26"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_27",
+          "name": "Y_27",
+          "description": "Benchmark target composition, component 27 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_27"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_28",
+          "name": "Y_28",
+          "description": "Benchmark target composition, component 28 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_28"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_29",
+          "name": "Y_29",
+          "description": "Benchmark target composition, component 29 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_29"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_30",
+          "name": "Y_30",
+          "description": "Benchmark target composition, component 30 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_30"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_31",
+          "name": "Y_31",
+          "description": "Benchmark target composition, component 31 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_31"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_32",
+          "name": "Y_32",
+          "description": "Benchmark target composition, component 32 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_32"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_33",
+          "name": "Y_33",
+          "description": "Benchmark target composition, component 33 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_33"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_34",
+          "name": "Y_34",
+          "description": "Benchmark target composition, component 34 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_34"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_35",
+          "name": "Y_35",
+          "description": "Benchmark target composition, component 35 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_35"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_36",
+          "name": "Y_36",
+          "description": "Benchmark target composition, component 36 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_36"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_37",
+          "name": "Y_37",
+          "description": "Benchmark target composition, component 37 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_37"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_38",
+          "name": "Y_38",
+          "description": "Benchmark target composition, component 38 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_38"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_39",
+          "name": "Y_39",
+          "description": "Benchmark target composition, component 39 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_39"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_40",
+          "name": "Y_40",
+          "description": "Benchmark target composition, component 40 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_40"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_41",
+          "name": "Y_41",
+          "description": "Benchmark target composition, component 41 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_41"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_42",
+          "name": "Y_42",
+          "description": "Benchmark target composition, component 42 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_42"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_43",
+          "name": "Y_43",
+          "description": "Benchmark target composition, component 43 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_43"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_44",
+          "name": "Y_44",
+          "description": "Benchmark target composition, component 44 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_44"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_45",
+          "name": "Y_45",
+          "description": "Benchmark target composition, component 45 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_45"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_46",
+          "name": "Y_46",
+          "description": "Benchmark target composition, component 46 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_46"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_47",
+          "name": "Y_47",
+          "description": "Benchmark target composition, component 47 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_47"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_48",
+          "name": "Y_48",
+          "description": "Benchmark target composition, component 48 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_48"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_49",
+          "name": "Y_49",
+          "description": "Benchmark target composition, component 49 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_49"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sigma_true",
+          "name": "sigma_true",
+          "description": "True local scale from the synthetic generator.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sigma_true"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "Benchmark-scale exact methods are too expensive here; the paper marks the exact-reference entries separately as D6 dagger cells. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "Synthetic diagnostic design deliberately selects one heterogeneity pattern; it does not sample a human population or establish demographic representativeness.",
+  "rai:personalSensitiveInformation": "Generated numerical simulations; no human-subject records or personal attributes are used in these synthetic arrays.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper’s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": true,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D6.yaml"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D6/R.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D6/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D6/Y.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D6/sigma_true.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-d6_high_k",
+    "description": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D6.yaml. Configuration:\nexperiment: d6_high_k\n\ndgp:\n  name: high_k\n  K: 50\n  sigma_min: 0.05\n  c: 0.35\n  d_x: 10\n\ndata:\n  n_train: 5000\n  n_cal: 5000\n  n_scale_est: 2500\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - partition\n  - twostage\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: entropy\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D6.yaml. Configuration:\nexperiment: d6_high_k\n\ndgp:\n  name: high_k\n  K: 50\n  sigma_min: 0.05\n  c: 0.35\n  d_x: 10\n\ndata:\n  n_train: 5000\n  n_cal: 5000\n  n_scale_est: 2500\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - partition\n  - twostage\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: entropy\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}

croissant/pbmc3k_pseudobulk.json

@@ -0,0 +1,422 @@
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/pbmc3k_pseudobulk/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#pbmc3k_pseudobulk",
+  "name": "SimplexTasks-12 — PBMC3K pseudobulk deconvolution",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/derived/pbmc_pseudobulk",
+  "description": "PBMC3K pseudobulk deconvolution is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 5000 samples with 8 simplex components. Frozen predictor: NNLS deconvolution from pseudobulk mixtures. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/pbmc_pseudobulk.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "pbmc3k_pseudobulk"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "pbmc3k_pseudobulk.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/pbmc3k_pseudobulk.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "241810 B",
+      "sha256": "db000c2f89ac6fcc5a599d518f9a2ed1b1aeff08ab09a456b2658dcf5e2f8eee"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "5000 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_3",
+          "name": "U_3",
+          "description": "Frozen predictor output, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_4",
+          "name": "U_4",
+          "description": "Frozen predictor output, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_5",
+          "name": "U_5",
+          "description": "Frozen predictor output, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_6",
+          "name": "U_6",
+          "description": "Frozen predictor output, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_7",
+          "name": "U_7",
+          "description": "Frozen predictor output, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_3",
+          "name": "Y_3",
+          "description": "Benchmark target composition, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_4",
+          "name": "Y_4",
+          "description": "Benchmark target composition, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_5",
+          "name": "Y_5",
+          "description": "Benchmark target composition, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_6",
+          "name": "Y_6",
+          "description": "Benchmark target composition, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_7",
+          "name": "Y_7",
+          "description": "Benchmark target composition, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_7"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "This is a semi-synthetic control rather than a deployment-ready deconvolution benchmark. The grouped-repair story is sensitive to the pseudobulk concentration setting, which is documented in the paper appendix. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "A limited single-cell reference and synthetic mixture choices restrict donor, population, cell-state and laboratory diversity.",
+  "rai:personalSensitiveInformation": "Derived data originate from human blood single-cell reference data. Synthetic cell-type fractions and deconvolution predictions are released; they are not clinical patient records and are not intended for individual identification.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper’s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": true,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://www.10xgenomics.com/datasets/3-k-pbm-cs-from-a-healthy-donor-1-standard-1-1-0"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/pbmc_pseudobulk/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/pbmc_pseudobulk/Y.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-pbmc3k_pseudobulk",
+    "description": "Pseudobulk cell-type fractions generated from PBMC3K together with a fixed NNLS deconvolution predictor. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. This package exports pseudobulk-derived arrays and signatures rather than the raw single-cell matrix. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/pbmc_pseudobulk.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Pseudobulk cell-type fractions generated from PBMC3K together with a fixed NNLS deconvolution predictor. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. This package exports pseudobulk-derived arrays and signatures rather than the raw single-cell matrix. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/pbmc_pseudobulk.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}

croissant/samson_unmixing.json

@@ -0,0 +1,272 @@
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/samson_unmixing/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#samson_unmixing",
+  "name": "SimplexTasks-12 — Samson hyperspectral unmixing",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/derived/samson_unmixing",
+  "description": "Samson hyperspectral unmixing is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 9025 samples with 3 simplex components. Frozen predictor: NMF abundance estimator. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/samson_unmixing.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "samson_unmixing"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "samson_unmixing.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/samson_unmixing.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "245140 B",
+      "sha256": "7be9f1cb90802b87866628afe8bab3d7c58e1671695e9497e9e1100dda7fa651"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "9025 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "The benchmark files are public and source-cited, but the upstream bundle did not include an explicit license file; downstream reuse should preserve attribution and the original benchmark context. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "One scene and three endmembers restrict geographic, material and sensor diversity; results may depend on spatial structure.",
+  "rai:personalSensitiveInformation": "The released arrays describe material abundances in a hyperspectral scene; the task documentation identifies no personal attributes in U and Y.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper’s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": false,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://rslab.ut.ac.ir/data"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/samson_unmixing/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/samson_unmixing/Y.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-samson_unmixing",
+    "description": "Three-endmember abundance compositions paired with a frozen NMF abundance estimator. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. The public bundle did not ship an explicit license file; keep attribution with any downstream reuse. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/samson_unmixing.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Three-endmember abundance compositions paired with a frozen NMF abundance estimator. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. The public bundle did not ship an explicit license file; keep attribution with any downstream reuse. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/samson_unmixing.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}

croissant/topics_20ng.json

@@ -0,0 +1,482 @@
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/topics_20ng/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#topics_20ng",
+  "name": "SimplexTasks-12 — 20 Newsgroups topics",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/derived/topics_20newsgroups",
+  "description": "20 Newsgroups topics is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 7539 samples with 10 simplex components. Frozen predictor: TF-IDF to topic-mixture kNN regressor. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/topics_20newsgroups.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "topics_20ng"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "topics_20ng.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/topics_20ng.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "784025 B",
+      "sha256": "67832ac04d380db18c42ddde4cd56909d5d8644e975c57885078fcbdbe8cb604"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "7539 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_3",
+          "name": "U_3",
+          "description": "Frozen predictor output, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_4",
+          "name": "U_4",
+          "description": "Frozen predictor output, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_5",
+          "name": "U_5",
+          "description": "Frozen predictor output, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_6",
+          "name": "U_6",
+          "description": "Frozen predictor output, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_7",
+          "name": "U_7",
+          "description": "Frozen predictor output, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_8",
+          "name": "U_8",
+          "description": "Frozen predictor output, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_9",
+          "name": "U_9",
+          "description": "Frozen predictor output, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_9"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_3",
+          "name": "Y_3",
+          "description": "Benchmark target composition, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_4",
+          "name": "Y_4",
+          "description": "Benchmark target composition, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_5",
+          "name": "Y_5",
+          "description": "Benchmark target composition, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_6",
+          "name": "Y_6",
+          "description": "Benchmark target composition, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_7",
+          "name": "Y_7",
+          "description": "Benchmark target composition, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_8",
+          "name": "Y_8",
+          "description": "Benchmark target composition, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_9",
+          "name": "Y_9",
+          "description": "Benchmark target composition, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_9"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "Both the target and the predictor output are model-derived topic mixtures. The release exposes derived simplex arrays, not the raw text corpus. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "English newsgroup selection and model-generated topic targets can encode language, historical, topic and model biases.",
+  "rai:personalSensitiveInformation": "The source corpus can contain names, email addresses, political and religious discussions; the released U and Y are derived topic mixtures, without raw text. No formal privacy guarantee is claimed.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper’s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": false,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "http://qwone.com/~jason/20Newsgroups/"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/topics_20newsgroups/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/topics_20newsgroups/Y.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-topics_20ng",
+    "description": "Fixed 10-topic compositions constructed once from the 20 Newsgroups corpus, with a TF-IDF-to-topic-mixture kNN regressor as the predictor. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. This release exposes only derived simplex arrays, not the raw text corpus. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/topics_20newsgroups.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Fixed 10-topic compositions constructed once from the 20 Newsgroups corpus, with a TF-IDF-to-topic-mixture kNN regressor as the predictor. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. This release exposes only derived simplex arrays, not the raw text corpus. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/topics_20newsgroups.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}

croissant/utkface_age_ldl.json

@@ -0,0 +1,482 @@
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/utkface_age_ldl/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#utkface_age_ldl",
+  "name": "SimplexTasks-12 — UTKFace age label distributions",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/derived/utkface_ldl",
+  "description": "UTKFace age label distributions is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 23687 samples with 10 simplex components. Frozen predictor: Thumbnail feature PCA+kNN age-distribution regressor. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/utkface_ldl.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "utkface_age_ldl"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "utkface_age_ldl.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/utkface_age_ldl.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "925580 B",
+      "sha256": "e6b2f9a64d6a2da09e247093225d15b3ababd0387235c2285ce7af27f63f1a07"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "23687 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_3",
+          "name": "U_3",
+          "description": "Frozen predictor output, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_4",
+          "name": "U_4",
+          "description": "Frozen predictor output, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_5",
+          "name": "U_5",
+          "description": "Frozen predictor output, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_6",
+          "name": "U_6",
+          "description": "Frozen predictor output, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_7",
+          "name": "U_7",
+          "description": "Frozen predictor output, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_8",
+          "name": "U_8",
+          "description": "Frozen predictor output, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_9",
+          "name": "U_9",
+          "description": "Frozen predictor output, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_9"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_3",
+          "name": "Y_3",
+          "description": "Benchmark target composition, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_4",
+          "name": "Y_4",
+          "description": "Benchmark target composition, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_5",
+          "name": "Y_5",
+          "description": "Benchmark target composition, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_6",
+          "name": "Y_6",
+          "description": "Benchmark target composition, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_7",
+          "name": "Y_7",
+          "description": "Benchmark target composition, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_8",
+          "name": "Y_8",
+          "description": "Benchmark target composition, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_9",
+          "name": "Y_9",
+          "description": "Benchmark target composition, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_9"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "The underlying face images are not redistributed and remain subject to the source dataset's research-use constraints. The benchmark ships derived features and simplex arrays only. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "Face-image collection and age labels may encode demographic imbalance and annotation errors; group-wise demographic validity has not been established.",
+  "rai:personalSensitiveInformation": "Age-related distributions are derived from face imagery. The source includes age, gender and race annotations. This release’s U and Y retain age-related information but exclude raw face images; no formal anonymisation guarantee is claimed.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper’s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": false,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://susanqq.github.io/UTKFace/"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/utkface_ldl/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/utkface_ldl/Y.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-utkface_age_ldl",
+    "description": "A 10-bin age label distribution constructed from UTKFace ages and predicted from thumbnail image features using PCA+kNN regression. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. The package omits raw face images and keeps only derived features and simplex arrays. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/utkface_ldl.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "A 10-bin age label distribution constructed from UTKFace ages and predicted from thumbnail image features using PCA+kNN regression. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. The package omits raw face images and keeps only derived features and simplex arrays. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/utkface_ldl.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}

docs/croissant_validation/README.md

@@ -0,0 +1,13 @@
+# Ready-to-upload SimplexTasks-12 Croissant metadata
+
+Validated on 2026-10-16 against the public, de-anonymised Hugging Face release.
+
+Upload `SimplexTasks-12_Croissant_VALIDATED.zip` to the OpenReview Croissant File field. It contains 12 individual JSON files, one per task. Dataset URL: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data
+
+All 12 files pass all four official online checks: JSON format, Croissant schema, required Responsible AI metadata, and record generation. Reports are saved in validator_reports/. The official record-generation test checks that a record can be generated; it is not a scientific or legal certification.
+
+The public dataset now includes 12 lossless Parquet tables. Every downloaded table was checked without authentication against the original frozen NumPy arrays for exact value, dtype and row-order equality. Sources remain unchanged. The metadata points to an immutable data commit, preventing later repository edits from silently changing validation inputs.
+
+The dataset card license category is now `other`, matching CITATION.cff and the existing task-specific artifact license notice. Metadata links this notice and retains upstream source restrictions; no new blanket license or third-party rights are granted.
+
+The metadata and validation reports are also published in the Hugging Face repository. The previous NumPy-based ZIP is superseded by this package.

docs/croissant_validation/affectivetext_emotions.md

@@ -0,0 +1,383 @@
+# CROISSANT VALIDATION REPORT
+================================================================================
+## VALIDATION RESULTS
+--------------------------------------------------------------------------------
+Starting validation for file: affectivetext_emotions.json
+### JSON Format Validation
+✓
+The file is valid JSON.
+### Croissant Schema Validation
+✓
+The dataset passes Croissant validation.
+### Responsible AI Metadata
+✓
+All required Responsible AI metadata fields are present.
+### Records Generation Test
+✓
+Record set 'https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/affectivetext_emotions/samples' passed validation.
+## JSON-LD REFERENCE
+================================================================================
+```json
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/affectivetext_emotions/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#affectivetext_emotions",
+  "name": "SimplexTasks-12 \u2014 SemEval AffectiveText emotions",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/derived/affectivetext",
+  "description": "SemEval AffectiveText emotions is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 1000 samples with 6 simplex components. Frozen predictor: Frozen zero-shot emotion scorer. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/affectivetext.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "affectivetext_emotions"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "affectivetext_emotions.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/affectivetext_emotions.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "35526 B",
+      "sha256": "14285f9b95857ff2bb91b005df1d3763084a2b3b91408bc6bdf92e577882d7b0"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "1000 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_3",
+          "name": "U_3",
+          "description": "Frozen predictor output, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_4",
+          "name": "U_4",
+          "description": "Frozen predictor output, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_5",
+          "name": "U_5",
+          "description": "Frozen predictor output, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_3",
+          "name": "Y_3",
+          "description": "Benchmark target composition, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_4",
+          "name": "Y_4",
+          "description": "Benchmark target composition, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_5",
+          "name": "Y_5",
+          "description": "Benchmark target composition, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_5"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "The reported benchmark tables use a frozen closed-model cache; the release omits raw headlines and raw API responses. A fully open TF-IDF+SVD+kNN fallback cache-builder is included for local reproduction, but it is not identical to the main predictor used in the paper tables. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "English headline selection, subjective emotion labels, and the frozen model may introduce language, cultural and annotation biases. This task does not measure an individual\u2019s emotional state.",
+  "rai:personalSensitiveInformation": "Derived emotion mixtures and headline identifiers are released; raw headlines and raw API responses are excluded. Emotion scores describe headlines, not diagnosed mental states.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper\u2019s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": false,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://aclanthology.org/S07-1013/"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/affectivetext/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/affectivetext/Y.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-affectivetext_emotions",
+    "description": "Normalized six-emotion gold scores from SemEval-2007 AffectiveText paired with a frozen zero-shot scorer cache. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. The package omits raw headlines and raw API responses, and keeps only derived score arrays. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/affectivetext.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Normalized six-emotion gold scores from SemEval-2007 AffectiveText paired with a frozen zero-shot scorer cache. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. The package omits raw headlines and raw API responses, and keeps only derived score arrays. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/affectivetext.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}
+```

docs/croissant_validation/cifar10_softmax.md

@@ -0,0 +1,503 @@
+# CROISSANT VALIDATION REPORT
+================================================================================
+## VALIDATION RESULTS
+--------------------------------------------------------------------------------
+Starting validation for file: cifar10_softmax.json
+### JSON Format Validation
+✓
+The file is valid JSON.
+### Croissant Schema Validation
+✓
+The dataset passes Croissant validation.
+### Responsible AI Metadata
+✓
+All required Responsible AI metadata fields are present.
+### Records Generation Test
+✓
+Record set 'https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/cifar10_softmax/samples' passed validation.
+## JSON-LD REFERENCE
+================================================================================
+```json
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/cifar10_softmax/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#cifar10_softmax",
+  "name": "SimplexTasks-12 \u2014 CIFAR-10 softmax",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/derived/cifar10_softmax",
+  "description": "CIFAR-10 softmax is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 10000 samples with 10 simplex components. Frozen predictor: Frozen ResNet-18 softmax cache. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/cifar10_softmax.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "cifar10_softmax"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "cifar10_softmax.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/cifar10_softmax.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "590269 B",
+      "sha256": "23d39b4db178de94ee8b76d8c64e0817ca0c9404a5bfec98369b68320b0ef5dc"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "10000 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_3",
+          "name": "U_3",
+          "description": "Frozen predictor output, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_4",
+          "name": "U_4",
+          "description": "Frozen predictor output, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_5",
+          "name": "U_5",
+          "description": "Frozen predictor output, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_6",
+          "name": "U_6",
+          "description": "Frozen predictor output, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_7",
+          "name": "U_7",
+          "description": "Frozen predictor output, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_8",
+          "name": "U_8",
+          "description": "Frozen predictor output, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_9",
+          "name": "U_9",
+          "description": "Frozen predictor output, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_9"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_3",
+          "name": "Y_3",
+          "description": "Benchmark target composition, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_4",
+          "name": "Y_4",
+          "description": "Benchmark target composition, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_5",
+          "name": "Y_5",
+          "description": "Benchmark target composition, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_6",
+          "name": "Y_6",
+          "description": "Benchmark target composition, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_7",
+          "name": "Y_7",
+          "description": "Benchmark target composition, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_8",
+          "name": "Y_8",
+          "description": "Benchmark target composition, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_9",
+          "name": "Y_9",
+          "description": "Benchmark target composition, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_9"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "This is a classification proxy rather than a naturally continuous composition task. The release ships derived arrays only and does not mirror raw CIFAR-10 images. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "The ten selected image classes, image collection and frozen ResNet-18 may introduce selection and model biases; this is not representative of all visual recognition settings.",
+  "rai:personalSensitiveInformation": "Released U and Y contain class probabilities and one-hot targets; raw images are excluded. No direct personal identifiers are described in these arrays.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper\u2019s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": false,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://www.cs.toronto.edu/~kriz/cifar.html"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/cifar10_softmax/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/cifar10_softmax/Y.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-cifar10_softmax",
+    "description": "A 10-way one-hot class distribution paired with a frozen ResNet-18 softmax cache. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. No raw images are redistributed in this package. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/cifar10_softmax.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "A 10-way one-hot class distribution paired with a frozen ResNet-18 softmax cache. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. No raw images are redistributed in this package. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/cifar10_softmax.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}
+```

docs/croissant_validation/d1_homogeneous.md

@@ -0,0 +1,329 @@
+# CROISSANT VALIDATION REPORT
+================================================================================
+## VALIDATION RESULTS
+--------------------------------------------------------------------------------
+Starting validation for file: d1_homogeneous.json
+### JSON Format Validation
+✓
+The file is valid JSON.
+### Croissant Schema Validation
+✓
+The dataset passes Croissant validation.
+### Responsible AI Metadata
+✓
+All required Responsible AI metadata fields are present.
+### Records Generation Test
+✓
+Record set 'https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d1_homogeneous/samples' passed validation.
+## JSON-LD REFERENCE
+================================================================================
+```json
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d1_homogeneous/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#d1_homogeneous",
+  "name": "SimplexTasks-12 \u2014 d1 homogeneous",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/synthetic/D1",
+  "description": "d1 homogeneous is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 7500 samples with 3 simplex components. Synthetic regime: Homogeneous negative control. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/synthetic/D1_homogeneous.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "d1_homogeneous"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "d1_homogeneous.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/d1_homogeneous.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "295831 B",
+      "sha256": "9fed050352b82559a340d492cb51dbf3f554f15cc0724b004df45319dee38d49"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "7500 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/R",
+          "name": "R",
+          "description": "Frozen residual score.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "R"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sigma_true",
+          "name": "sigma_true",
+          "description": "True local scale from the synthetic generator.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sigma_true"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "Intended to confirm that the benchmark does not manufacture allocation failures when the DGP is homogeneous. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "Synthetic diagnostic design deliberately selects one heterogeneity pattern; it does not sample a human population or establish demographic representativeness.",
+  "rai:personalSensitiveInformation": "Generated numerical simulations; no human-subject records or personal attributes are used in these synthetic arrays.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper\u2019s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": true,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D1.yaml"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D1/R.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D1/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D1/Y.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D1/sigma_true.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-d1_homogeneous",
+    "description": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D1.yaml. Configuration:\nexperiment: d1_homogeneous\n\ndgp:\n  name: pure_scale\n  K: 3\n  sigma_min: 0.2\n  c: 0.0\n  d_x: 2\n\ndata:\n  n_train: 1000\n  n_cal: 1000\n  n_scale_est: 500\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - partition\n  - twostage\n  - oneshot\n  - trainres\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D1.yaml. Configuration:\nexperiment: d1_homogeneous\n\ndgp:\n  name: pure_scale\n  K: 3\n  sigma_min: 0.2\n  c: 0.0\n  d_x: 2\n\ndata:\n  n_train: 1000\n  n_cal: 1000\n  n_scale_est: 500\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - partition\n  - twostage\n  - oneshot\n  - trainres\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}
+```

docs/croissant_validation/d2_pure_scale.md

@@ -0,0 +1,329 @@
+# CROISSANT VALIDATION REPORT
+================================================================================
+## VALIDATION RESULTS
+--------------------------------------------------------------------------------
+Starting validation for file: d2_pure_scale.json
+### JSON Format Validation
+✓
+The file is valid JSON.
+### Croissant Schema Validation
+✓
+The dataset passes Croissant validation.
+### Responsible AI Metadata
+✓
+All required Responsible AI metadata fields are present.
+### Records Generation Test
+✓
+Record set 'https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d2_pure_scale/samples' passed validation.
+## JSON-LD REFERENCE
+================================================================================
+```json
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d2_pure_scale/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#d2_pure_scale",
+  "name": "SimplexTasks-12 \u2014 d2 pure scale",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/synthetic/D2",
+  "description": "d2 pure scale is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 6250 samples with 3 simplex components. Synthetic regime: Smooth scale heterogeneity. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/synthetic/D2_smooth_scale.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "d2_pure_scale"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "d2_pure_scale.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/d2_pure_scale.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "278811 B",
+      "sha256": "de27d8fa6b0841f45fc9336fa717423d36e5f605145bc7001ee78e8076b52c2a"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "6250 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/R",
+          "name": "R",
+          "description": "Frozen residual score.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "R"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sigma_true",
+          "name": "sigma_true",
+          "description": "True local scale from the synthetic generator.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sigma_true"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "This regime isolates scale variation only; it is not intended to cover bias-type misspecification or changing tail shape. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "Synthetic diagnostic design deliberately selects one heterogeneity pattern; it does not sample a human population or establish demographic representativeness.",
+  "rai:personalSensitiveInformation": "Generated numerical simulations; no human-subject records or personal attributes are used in these synthetic arrays.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper\u2019s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": true,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D2.yaml"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D2/R.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D2/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D2/Y.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D2/sigma_true.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-d2_pure_scale",
+    "description": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D2.yaml. Configuration:\nexperiment: d2_pure_scale\n\ndgp:\n  name: pure_scale\n  K: 3\n  sigma_min: 0.1\n  c: 0.5\n  d_x: 2\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - fullcp\n  - jackknife_plus\n  - partition\n  - twostage\n  - oneshot\n  - trainres\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D2.yaml. Configuration:\nexperiment: d2_pure_scale\n\ndgp:\n  name: pure_scale\n  K: 3\n  sigma_min: 0.1\n  c: 0.5\n  d_x: 2\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - fullcp\n  - jackknife_plus\n  - partition\n  - twostage\n  - oneshot\n  - trainres\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}
+```

docs/croissant_validation/d3_discrete_groups_aligned.md

@@ -0,0 +1,539 @@
+# CROISSANT VALIDATION REPORT
+================================================================================
+## VALIDATION RESULTS
+--------------------------------------------------------------------------------
+Starting validation for file: d3_discrete_groups_aligned.json
+### JSON Format Validation
+✓
+The file is valid JSON.
+### Croissant Schema Validation
+✓
+The dataset passes Croissant validation.
+### Responsible AI Metadata
+✓
+All required Responsible AI metadata fields are present.
+### Records Generation Test
+✓
+Record set 'https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d3_discrete_groups_aligned/samples' passed validation.
+## JSON-LD REFERENCE
+================================================================================
+```json
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d3_discrete_groups_aligned/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#d3_discrete_groups_aligned",
+  "name": "SimplexTasks-12 \u2014 d3 discrete groups aligned",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/synthetic/D3",
+  "description": "d3 discrete groups aligned is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 6250 samples with 10 simplex components. Synthetic regime: Aligned discrete groups. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/synthetic/D3_aligned_discrete.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "d3_discrete_groups_aligned"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "d3_discrete_groups_aligned.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/d3_discrete_groups_aligned.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "715419 B",
+      "sha256": "f6daa6cfb82fb63b206efb5d8211d200df9dc46552a9656d02d9fa7942344298"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "6250 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/R",
+          "name": "R",
+          "description": "Frozen residual score.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "R"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_3",
+          "name": "U_3",
+          "description": "Frozen predictor output, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_4",
+          "name": "U_4",
+          "description": "Frozen predictor output, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_5",
+          "name": "U_5",
+          "description": "Frozen predictor output, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_6",
+          "name": "U_6",
+          "description": "Frozen predictor output, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_7",
+          "name": "U_7",
+          "description": "Frozen predictor output, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_8",
+          "name": "U_8",
+          "description": "Frozen predictor output, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_9",
+          "name": "U_9",
+          "description": "Frozen predictor output, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_9"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_3",
+          "name": "Y_3",
+          "description": "Benchmark target composition, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_4",
+          "name": "Y_4",
+          "description": "Benchmark target composition, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_5",
+          "name": "Y_5",
+          "description": "Benchmark target composition, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_6",
+          "name": "Y_6",
+          "description": "Benchmark target composition, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_7",
+          "name": "Y_7",
+          "description": "Benchmark target composition, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_8",
+          "name": "Y_8",
+          "description": "Benchmark target composition, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_9",
+          "name": "Y_9",
+          "description": "Benchmark target composition, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_9"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sigma_true",
+          "name": "sigma_true",
+          "description": "True local scale from the synthetic generator.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sigma_true"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "The grouping is built into the DGP. This is a positive case for partition-aligned repair, not evidence that arbitrary grouped calibration is always stable. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "Synthetic diagnostic design deliberately selects one heterogeneity pattern; it does not sample a human population or establish demographic representativeness.",
+  "rai:personalSensitiveInformation": "Generated numerical simulations; no human-subject records or personal attributes are used in these synthetic arrays.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper\u2019s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": true,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D3.yaml"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D3/R.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D3/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D3/Y.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D3/sigma_true.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-d3_discrete_groups_aligned",
+    "description": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D3.yaml. Configuration:\nexperiment: d3_discrete_groups_aligned\n\ndgp:\n  name: discrete_groups\n  K: 10\n  sigma_low: 0.08\n  sigma_high: 0.30\n  d_x: 5\n  easy_classes: 5\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - partition\n  - twostage\n  - fullcp\n  - jackknife_plus\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: argmax_group\n  n_strata: 2\n  split_index: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D3.yaml. Configuration:\nexperiment: d3_discrete_groups_aligned\n\ndgp:\n  name: discrete_groups\n  K: 10\n  sigma_low: 0.08\n  sigma_high: 0.30\n  d_x: 5\n  easy_classes: 5\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - partition\n  - twostage\n  - fullcp\n  - jackknife_plus\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: argmax_group\n  n_strata: 2\n  split_index: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}
+```

docs/croissant_validation/d4_model_bias.md

@@ -0,0 +1,329 @@
+# CROISSANT VALIDATION REPORT
+================================================================================
+## VALIDATION RESULTS
+--------------------------------------------------------------------------------
+Starting validation for file: d4_model_bias.json
+### JSON Format Validation
+✓
+The file is valid JSON.
+### Croissant Schema Validation
+✓
+The dataset passes Croissant validation.
+### Responsible AI Metadata
+✓
+All required Responsible AI metadata fields are present.
+### Records Generation Test
+✓
+Record set 'https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d4_model_bias/samples' passed validation.
+## JSON-LD REFERENCE
+================================================================================
+```json
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d4_model_bias/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#d4_model_bias",
+  "name": "SimplexTasks-12 \u2014 d4 model bias",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/synthetic/D4",
+  "description": "d4 model bias is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 6250 samples with 3 simplex components. Synthetic regime: Bias-type heterogeneity. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/synthetic/D4_bias_type.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "d4_model_bias"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "d4_model_bias.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/d4_model_bias.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "278534 B",
+      "sha256": "dfab38720fb1903807b2c2e90008d8b672ffe682a9dd184f7383acd7cccfba0f"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "6250 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/R",
+          "name": "R",
+          "description": "Frozen residual score.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "R"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sigma_true",
+          "name": "sigma_true",
+          "description": "True local scale from the synthetic generator.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sigma_true"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "This regime is not reducible to a single local scale field, so normalization cannot be expected to fully repair it. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "Synthetic diagnostic design deliberately selects one heterogeneity pattern; it does not sample a human population or establish demographic representativeness.",
+  "rai:personalSensitiveInformation": "Generated numerical simulations; no human-subject records or personal attributes are used in these synthetic arrays.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper\u2019s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": true,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D4.yaml"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D4/R.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D4/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D4/Y.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D4/sigma_true.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-d4_model_bias",
+    "description": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D4.yaml. Configuration:\nexperiment: d4_model_bias\n\ndgp:\n  name: model_bias\n  K: 3\n  sigma_min: 0.1\n  c: 0.15\n  d_x: 2\n  bias_scale: 0.45\n  bias_type: rotational\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - fullcp\n  - jackknife_plus\n  - partition\n  - twostage\n  - oneshot\n  - trainres\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D4.yaml. Configuration:\nexperiment: d4_model_bias\n\ndgp:\n  name: model_bias\n  K: 3\n  sigma_min: 0.1\n  c: 0.15\n  d_x: 2\n  bias_scale: 0.45\n  bias_type: rotational\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - fullcp\n  - jackknife_plus\n  - partition\n  - twostage\n  - oneshot\n  - trainres\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}
+```

docs/croissant_validation/d5_heavy_tail.md

@@ -0,0 +1,329 @@
+# CROISSANT VALIDATION REPORT
+================================================================================
+## VALIDATION RESULTS
+--------------------------------------------------------------------------------
+Starting validation for file: d5_heavy_tail.json
+### JSON Format Validation
+✓
+The file is valid JSON.
+### Croissant Schema Validation
+✓
+The dataset passes Croissant validation.
+### Responsible AI Metadata
+✓
+All required Responsible AI metadata fields are present.
+### Records Generation Test
+✓
+Record set 'https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d5_heavy_tail/samples' passed validation.
+## JSON-LD REFERENCE
+================================================================================
+```json
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d5_heavy_tail/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#d5_heavy_tail",
+  "name": "SimplexTasks-12 \u2014 d5 heavy tail",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/synthetic/D5",
+  "description": "d5 heavy tail is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 6250 samples with 3 simplex components. Synthetic regime: Heavy-tail robustness. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/synthetic/D5_heavy_tail.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "d5_heavy_tail"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "d5_heavy_tail.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/d5_heavy_tail.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "278749 B",
+      "sha256": "dd7237032d30971d9097851d8b0484935618bb42c5eb84ed38e0f3e50d345a12"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "6250 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/R",
+          "name": "R",
+          "description": "Frozen residual score.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "R"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sigma_true",
+          "name": "sigma_true",
+          "description": "True local scale from the synthetic generator.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sigma_true"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "The regime is still synthetic and targeted; it does not span every possible deviation from the baseline score distribution. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "Synthetic diagnostic design deliberately selects one heterogeneity pattern; it does not sample a human population or establish demographic representativeness.",
+  "rai:personalSensitiveInformation": "Generated numerical simulations; no human-subject records or personal attributes are used in these synthetic arrays.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper\u2019s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": true,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D5.yaml"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D5/R.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D5/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D5/Y.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D5/sigma_true.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-d5_heavy_tail",
+    "description": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D5.yaml. Configuration:\nexperiment: d5_heavy_tail\n\ndgp:\n  name: heavy_tail\n  K: 3\n  sigma_min: 0.1\n  c: 0.5\n  d_x: 2\n  df: 3.0\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - fullcp\n  - jackknife_plus\n  - partition\n  - twostage\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D5.yaml. Configuration:\nexperiment: d5_heavy_tail\n\ndgp:\n  name: heavy_tail\n  K: 3\n  sigma_min: 0.1\n  c: 0.5\n  d_x: 2\n  df: 3.0\n\ndata:\n  n_train: 500\n  n_cal: 500\n  n_scale_est: 250\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - fullcp\n  - jackknife_plus\n  - partition\n  - twostage\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: boundary\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}
+```

docs/croissant_validation/d6_high_k.md

@@ -0,0 +1,1739 @@
+# CROISSANT VALIDATION REPORT
+================================================================================
+## VALIDATION RESULTS
+--------------------------------------------------------------------------------
+Starting validation for file: d6_high_k.json
+### JSON Format Validation
+✓
+The file is valid JSON.
+### Croissant Schema Validation
+✓
+The dataset passes Croissant validation.
+### Responsible AI Metadata
+✓
+All required Responsible AI metadata fields are present.
+### Records Generation Test
+✓
+Record set 'https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d6_high_k/samples' passed validation.
+## JSON-LD REFERENCE
+================================================================================
+```json
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/d6_high_k/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#d6_high_k",
+  "name": "SimplexTasks-12 \u2014 d6 high k",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/synthetic/D6",
+  "description": "d6 high k is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 17500 samples with 50 simplex components. Synthetic regime: High-dimensional simplex. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/synthetic/D6_highK.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "d6_high_k"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "d6_high_k.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/d6_high_k.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "10028544 B",
+      "sha256": "2e4fe45eae47a67c9d3023d8b302dcc0f0b9b8c7e007f3ab07e2db6011e0bbfc"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "17500 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/R",
+          "name": "R",
+          "description": "Frozen residual score.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "R"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_3",
+          "name": "U_3",
+          "description": "Frozen predictor output, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_4",
+          "name": "U_4",
+          "description": "Frozen predictor output, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_5",
+          "name": "U_5",
+          "description": "Frozen predictor output, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_6",
+          "name": "U_6",
+          "description": "Frozen predictor output, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_7",
+          "name": "U_7",
+          "description": "Frozen predictor output, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_8",
+          "name": "U_8",
+          "description": "Frozen predictor output, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_9",
+          "name": "U_9",
+          "description": "Frozen predictor output, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_9"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_10",
+          "name": "U_10",
+          "description": "Frozen predictor output, component 10 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_10"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_11",
+          "name": "U_11",
+          "description": "Frozen predictor output, component 11 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_11"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_12",
+          "name": "U_12",
+          "description": "Frozen predictor output, component 12 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_12"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_13",
+          "name": "U_13",
+          "description": "Frozen predictor output, component 13 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_13"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_14",
+          "name": "U_14",
+          "description": "Frozen predictor output, component 14 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_14"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_15",
+          "name": "U_15",
+          "description": "Frozen predictor output, component 15 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_15"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_16",
+          "name": "U_16",
+          "description": "Frozen predictor output, component 16 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_16"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_17",
+          "name": "U_17",
+          "description": "Frozen predictor output, component 17 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_17"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_18",
+          "name": "U_18",
+          "description": "Frozen predictor output, component 18 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_18"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_19",
+          "name": "U_19",
+          "description": "Frozen predictor output, component 19 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_19"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_20",
+          "name": "U_20",
+          "description": "Frozen predictor output, component 20 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_20"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_21",
+          "name": "U_21",
+          "description": "Frozen predictor output, component 21 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_21"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_22",
+          "name": "U_22",
+          "description": "Frozen predictor output, component 22 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_22"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_23",
+          "name": "U_23",
+          "description": "Frozen predictor output, component 23 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_23"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_24",
+          "name": "U_24",
+          "description": "Frozen predictor output, component 24 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_24"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_25",
+          "name": "U_25",
+          "description": "Frozen predictor output, component 25 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_25"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_26",
+          "name": "U_26",
+          "description": "Frozen predictor output, component 26 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_26"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_27",
+          "name": "U_27",
+          "description": "Frozen predictor output, component 27 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_27"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_28",
+          "name": "U_28",
+          "description": "Frozen predictor output, component 28 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_28"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_29",
+          "name": "U_29",
+          "description": "Frozen predictor output, component 29 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_29"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_30",
+          "name": "U_30",
+          "description": "Frozen predictor output, component 30 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_30"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_31",
+          "name": "U_31",
+          "description": "Frozen predictor output, component 31 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_31"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_32",
+          "name": "U_32",
+          "description": "Frozen predictor output, component 32 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_32"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_33",
+          "name": "U_33",
+          "description": "Frozen predictor output, component 33 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_33"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_34",
+          "name": "U_34",
+          "description": "Frozen predictor output, component 34 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_34"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_35",
+          "name": "U_35",
+          "description": "Frozen predictor output, component 35 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_35"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_36",
+          "name": "U_36",
+          "description": "Frozen predictor output, component 36 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_36"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_37",
+          "name": "U_37",
+          "description": "Frozen predictor output, component 37 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_37"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_38",
+          "name": "U_38",
+          "description": "Frozen predictor output, component 38 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_38"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_39",
+          "name": "U_39",
+          "description": "Frozen predictor output, component 39 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_39"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_40",
+          "name": "U_40",
+          "description": "Frozen predictor output, component 40 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_40"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_41",
+          "name": "U_41",
+          "description": "Frozen predictor output, component 41 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_41"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_42",
+          "name": "U_42",
+          "description": "Frozen predictor output, component 42 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_42"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_43",
+          "name": "U_43",
+          "description": "Frozen predictor output, component 43 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_43"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_44",
+          "name": "U_44",
+          "description": "Frozen predictor output, component 44 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_44"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_45",
+          "name": "U_45",
+          "description": "Frozen predictor output, component 45 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_45"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_46",
+          "name": "U_46",
+          "description": "Frozen predictor output, component 46 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_46"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_47",
+          "name": "U_47",
+          "description": "Frozen predictor output, component 47 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_47"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_48",
+          "name": "U_48",
+          "description": "Frozen predictor output, component 48 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_48"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_49",
+          "name": "U_49",
+          "description": "Frozen predictor output, component 49 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_49"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_3",
+          "name": "Y_3",
+          "description": "Benchmark target composition, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_4",
+          "name": "Y_4",
+          "description": "Benchmark target composition, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_5",
+          "name": "Y_5",
+          "description": "Benchmark target composition, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_6",
+          "name": "Y_6",
+          "description": "Benchmark target composition, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_7",
+          "name": "Y_7",
+          "description": "Benchmark target composition, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_8",
+          "name": "Y_8",
+          "description": "Benchmark target composition, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_9",
+          "name": "Y_9",
+          "description": "Benchmark target composition, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_9"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_10",
+          "name": "Y_10",
+          "description": "Benchmark target composition, component 10 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_10"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_11",
+          "name": "Y_11",
+          "description": "Benchmark target composition, component 11 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_11"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_12",
+          "name": "Y_12",
+          "description": "Benchmark target composition, component 12 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_12"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_13",
+          "name": "Y_13",
+          "description": "Benchmark target composition, component 13 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_13"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_14",
+          "name": "Y_14",
+          "description": "Benchmark target composition, component 14 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_14"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_15",
+          "name": "Y_15",
+          "description": "Benchmark target composition, component 15 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_15"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_16",
+          "name": "Y_16",
+          "description": "Benchmark target composition, component 16 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_16"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_17",
+          "name": "Y_17",
+          "description": "Benchmark target composition, component 17 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_17"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_18",
+          "name": "Y_18",
+          "description": "Benchmark target composition, component 18 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_18"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_19",
+          "name": "Y_19",
+          "description": "Benchmark target composition, component 19 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_19"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_20",
+          "name": "Y_20",
+          "description": "Benchmark target composition, component 20 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_20"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_21",
+          "name": "Y_21",
+          "description": "Benchmark target composition, component 21 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_21"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_22",
+          "name": "Y_22",
+          "description": "Benchmark target composition, component 22 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_22"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_23",
+          "name": "Y_23",
+          "description": "Benchmark target composition, component 23 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_23"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_24",
+          "name": "Y_24",
+          "description": "Benchmark target composition, component 24 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_24"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_25",
+          "name": "Y_25",
+          "description": "Benchmark target composition, component 25 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_25"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_26",
+          "name": "Y_26",
+          "description": "Benchmark target composition, component 26 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_26"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_27",
+          "name": "Y_27",
+          "description": "Benchmark target composition, component 27 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_27"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_28",
+          "name": "Y_28",
+          "description": "Benchmark target composition, component 28 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_28"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_29",
+          "name": "Y_29",
+          "description": "Benchmark target composition, component 29 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_29"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_30",
+          "name": "Y_30",
+          "description": "Benchmark target composition, component 30 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_30"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_31",
+          "name": "Y_31",
+          "description": "Benchmark target composition, component 31 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_31"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_32",
+          "name": "Y_32",
+          "description": "Benchmark target composition, component 32 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_32"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_33",
+          "name": "Y_33",
+          "description": "Benchmark target composition, component 33 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_33"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_34",
+          "name": "Y_34",
+          "description": "Benchmark target composition, component 34 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_34"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_35",
+          "name": "Y_35",
+          "description": "Benchmark target composition, component 35 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_35"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_36",
+          "name": "Y_36",
+          "description": "Benchmark target composition, component 36 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_36"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_37",
+          "name": "Y_37",
+          "description": "Benchmark target composition, component 37 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_37"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_38",
+          "name": "Y_38",
+          "description": "Benchmark target composition, component 38 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_38"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_39",
+          "name": "Y_39",
+          "description": "Benchmark target composition, component 39 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_39"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_40",
+          "name": "Y_40",
+          "description": "Benchmark target composition, component 40 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_40"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_41",
+          "name": "Y_41",
+          "description": "Benchmark target composition, component 41 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_41"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_42",
+          "name": "Y_42",
+          "description": "Benchmark target composition, component 42 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_42"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_43",
+          "name": "Y_43",
+          "description": "Benchmark target composition, component 43 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_43"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_44",
+          "name": "Y_44",
+          "description": "Benchmark target composition, component 44 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_44"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_45",
+          "name": "Y_45",
+          "description": "Benchmark target composition, component 45 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_45"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_46",
+          "name": "Y_46",
+          "description": "Benchmark target composition, component 46 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_46"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_47",
+          "name": "Y_47",
+          "description": "Benchmark target composition, component 47 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_47"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_48",
+          "name": "Y_48",
+          "description": "Benchmark target composition, component 48 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_48"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_49",
+          "name": "Y_49",
+          "description": "Benchmark target composition, component 49 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_49"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sigma_true",
+          "name": "sigma_true",
+          "description": "True local scale from the synthetic generator.",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sigma_true"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "Benchmark-scale exact methods are too expensive here; the paper marks the exact-reference entries separately as D6 dagger cells. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "Synthetic diagnostic design deliberately selects one heterogeneity pattern; it does not sample a human population or establish demographic representativeness.",
+  "rai:personalSensitiveInformation": "Generated numerical simulations; no human-subject records or personal attributes are used in these synthetic arrays.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper\u2019s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": true,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D6.yaml"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D6/R.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D6/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D6/Y.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/synthetic/D6/sigma_true.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-d6_high_k",
+    "description": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D6.yaml. Configuration:\nexperiment: d6_high_k\n\ndgp:\n  name: high_k\n  K: 50\n  sigma_min: 0.05\n  c: 0.35\n  d_x: 10\n\ndata:\n  n_train: 5000\n  n_cal: 5000\n  n_scale_est: 2500\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - partition\n  - twostage\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: entropy\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Generated numerical samples with seed 2026; generator and parameters are specified in https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/configs/synthetic/D6.yaml. Configuration:\nexperiment: d6_high_k\n\ndgp:\n  name: high_k\n  K: 50\n  sigma_min: 0.05\n  c: 0.35\n  d_x: 10\n\ndata:\n  n_train: 5000\n  n_cal: 5000\n  n_scale_est: 2500\n  n_test: 5000\n  n_rep: 200\n\nmethods:\n  - global\n  - partition\n  - twostage\n  - weighted\n  - oracle\n\nevaluation:\n  alpha: 0.1\n  strata_method: entropy\n  n_strata: 5\n\nweighting:\n  mode: inverse_sigma\n  source: knn_loo\n\nseed: 2026\n\nFrozen arrays were packaged by the SimplexUQ authors. No human annotation was used. The released arrays are one frozen sample pool; the experiment configuration also specifies repeated experiments. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}
+```

docs/croissant_validation/online_validation_summary.json

@@ -0,0 +1,102 @@
+{
+  "validator": "https://huggingface.co/spaces/JoaquinVanschoren/croissant-checker",
+  "validated_on": "2026-10-16",
+  "results": [
+    {
+      "task": "affectivetext_emotions",
+      "json": "pass",
+      "croissant_schema": "pass",
+      "required_rai": "pass",
+      "record_generation": "pass",
+      "metadata_sha256": "74ace383bbd207a0ecb55d035e64394139946c3c07d6d75ea2025cc5d2597d39"
+    },
+    {
+      "task": "cifar10_softmax",
+      "json": "pass",
+      "croissant_schema": "pass",
+      "required_rai": "pass",
+      "record_generation": "pass",
+      "metadata_sha256": "2c6d8c35f963232a7ca7bfeb4dd07248a2323924b9ed337d101a50e386daf6d0"
+    },
+    {
+      "task": "d1_homogeneous",
+      "json": "pass",
+      "croissant_schema": "pass",
+      "required_rai": "pass",
+      "record_generation": "pass",
+      "metadata_sha256": "28f33bfbb0516e1c5314bcf4ee3398ec1e3ad345e3b126c4a4d1dbd504b518e4"
+    },
+    {
+      "task": "d2_pure_scale",
+      "json": "pass",
+      "croissant_schema": "pass",
+      "required_rai": "pass",
+      "record_generation": "pass",
+      "metadata_sha256": "4c56227311eecdf9e6a10dd76f444fb37c2226c16f8f506df2c982f9fdd063cf"
+    },
+    {
+      "task": "d3_discrete_groups_aligned",
+      "json": "pass",
+      "croissant_schema": "pass",
+      "required_rai": "pass",
+      "record_generation": "pass",
+      "metadata_sha256": "44b2a015a75735de5f975c6b3e7496a934697c54232e3eae5fa4d51e39bf882e"
+    },
+    {
+      "task": "d4_model_bias",
+      "json": "pass",
+      "croissant_schema": "pass",
+      "required_rai": "pass",
+      "record_generation": "pass",
+      "metadata_sha256": "be3a2cc19b93cef7662f8a3904b653a6d33a3d252bee17e41940df0f28384b4e"
+    },
+    {
+      "task": "d5_heavy_tail",
+      "json": "pass",
+      "croissant_schema": "pass",
+      "required_rai": "pass",
+      "record_generation": "pass",
+      "metadata_sha256": "edd06abb97a93e716aacd303350820596cac9b7f48a08ae75d68af5f10425105"
+    },
+    {
+      "task": "d6_high_k",
+      "json": "pass",
+      "croissant_schema": "pass",
+      "required_rai": "pass",
+      "record_generation": "pass",
+      "metadata_sha256": "ed4c08f9e2fd7ec1d864631222b397cba5430f1acbdd8a0b51bc42e97518a774"
+    },
+    {
+      "task": "pbmc3k_pseudobulk",
+      "json": "pass",
+      "croissant_schema": "pass",
+      "required_rai": "pass",
+      "record_generation": "pass",
+      "metadata_sha256": "ccdab623b23d11e16bb2e7378be8c57f8d4a94f2e6bfd94da91a2980b31c20c6"
+    },
+    {
+      "task": "samson_unmixing",
+      "json": "pass",
+      "croissant_schema": "pass",
+      "required_rai": "pass",
+      "record_generation": "pass",
+      "metadata_sha256": "9b503b6dd8214e03b89041ac44239d1c3eb32052da1a2eb2b05a0244f82ecf0d"
+    },
+    {
+      "task": "topics_20ng",
+      "json": "pass",
+      "croissant_schema": "pass",
+      "required_rai": "pass",
+      "record_generation": "pass",
+      "metadata_sha256": "a3150369834fea3aa3503d4f7df0feb5fe67e00fdc838d36569e6d66f246a5a5"
+    },
+    {
+      "task": "utkface_age_ldl",
+      "json": "pass",
+      "croissant_schema": "pass",
+      "required_rai": "pass",
+      "record_generation": "pass",
+      "metadata_sha256": "9a10b285a041208c900d142a4dcb2a668ea09b08850d9f77dbefe2991de1d1a6"
+    }
+  ]
+}

docs/croissant_validation/pbmc3k_pseudobulk.md

@@ -0,0 +1,443 @@
+# CROISSANT VALIDATION REPORT
+================================================================================
+## VALIDATION RESULTS
+--------------------------------------------------------------------------------
+Starting validation for file: pbmc3k_pseudobulk.json
+### JSON Format Validation
+✓
+The file is valid JSON.
+### Croissant Schema Validation
+✓
+The dataset passes Croissant validation.
+### Responsible AI Metadata
+✓
+All required Responsible AI metadata fields are present.
+### Records Generation Test
+✓
+Record set 'https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/pbmc3k_pseudobulk/samples' passed validation.
+## JSON-LD REFERENCE
+================================================================================
+```json
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/pbmc3k_pseudobulk/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#pbmc3k_pseudobulk",
+  "name": "SimplexTasks-12 \u2014 PBMC3K pseudobulk deconvolution",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/derived/pbmc_pseudobulk",
+  "description": "PBMC3K pseudobulk deconvolution is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 5000 samples with 8 simplex components. Frozen predictor: NNLS deconvolution from pseudobulk mixtures. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/pbmc_pseudobulk.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "pbmc3k_pseudobulk"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "pbmc3k_pseudobulk.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/pbmc3k_pseudobulk.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "241810 B",
+      "sha256": "db000c2f89ac6fcc5a599d518f9a2ed1b1aeff08ab09a456b2658dcf5e2f8eee"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "5000 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_3",
+          "name": "U_3",
+          "description": "Frozen predictor output, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_4",
+          "name": "U_4",
+          "description": "Frozen predictor output, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_5",
+          "name": "U_5",
+          "description": "Frozen predictor output, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_6",
+          "name": "U_6",
+          "description": "Frozen predictor output, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_7",
+          "name": "U_7",
+          "description": "Frozen predictor output, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_3",
+          "name": "Y_3",
+          "description": "Benchmark target composition, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_4",
+          "name": "Y_4",
+          "description": "Benchmark target composition, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_5",
+          "name": "Y_5",
+          "description": "Benchmark target composition, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_6",
+          "name": "Y_6",
+          "description": "Benchmark target composition, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_7",
+          "name": "Y_7",
+          "description": "Benchmark target composition, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_7"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "This is a semi-synthetic control rather than a deployment-ready deconvolution benchmark. The grouped-repair story is sensitive to the pseudobulk concentration setting, which is documented in the paper appendix. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "A limited single-cell reference and synthetic mixture choices restrict donor, population, cell-state and laboratory diversity.",
+  "rai:personalSensitiveInformation": "Derived data originate from human blood single-cell reference data. Synthetic cell-type fractions and deconvolution predictions are released; they are not clinical patient records and are not intended for individual identification.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper\u2019s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": true,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://www.10xgenomics.com/datasets/3-k-pbm-cs-from-a-healthy-donor-1-standard-1-1-0"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/pbmc_pseudobulk/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/pbmc_pseudobulk/Y.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-pbmc3k_pseudobulk",
+    "description": "Pseudobulk cell-type fractions generated from PBMC3K together with a fixed NNLS deconvolution predictor. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. This package exports pseudobulk-derived arrays and signatures rather than the raw single-cell matrix. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/pbmc_pseudobulk.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Pseudobulk cell-type fractions generated from PBMC3K together with a fixed NNLS deconvolution predictor. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. This package exports pseudobulk-derived arrays and signatures rather than the raw single-cell matrix. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/pbmc_pseudobulk.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}
+```

docs/croissant_validation/public_data_verification.json

@@ -0,0 +1,90 @@
+{
+  "revision": "ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501",
+  "authentication": "none",
+  "results": [
+    {
+      "task": "affectivetext_emotions",
+      "public_download": "pass",
+      "sha256": "pass",
+      "all_values_and_dtypes": "pass",
+      "rows": 1000
+    },
+    {
+      "task": "cifar10_softmax",
+      "public_download": "pass",
+      "sha256": "pass",
+      "all_values_and_dtypes": "pass",
+      "rows": 10000
+    },
+    {
+      "task": "pbmc3k_pseudobulk",
+      "public_download": "pass",
+      "sha256": "pass",
+      "all_values_and_dtypes": "pass",
+      "rows": 5000
+    },
+    {
+      "task": "samson_unmixing",
+      "public_download": "pass",
+      "sha256": "pass",
+      "all_values_and_dtypes": "pass",
+      "rows": 9025
+    },
+    {
+      "task": "topics_20ng",
+      "public_download": "pass",
+      "sha256": "pass",
+      "all_values_and_dtypes": "pass",
+      "rows": 7539
+    },
+    {
+      "task": "utkface_age_ldl",
+      "public_download": "pass",
+      "sha256": "pass",
+      "all_values_and_dtypes": "pass",
+      "rows": 23687
+    },
+    {
+      "task": "d1_homogeneous",
+      "public_download": "pass",
+      "sha256": "pass",
+      "all_values_and_dtypes": "pass",
+      "rows": 7500
+    },
+    {
+      "task": "d2_pure_scale",
+      "public_download": "pass",
+      "sha256": "pass",
+      "all_values_and_dtypes": "pass",
+      "rows": 6250
+    },
+    {
+      "task": "d3_discrete_groups_aligned",
+      "public_download": "pass",
+      "sha256": "pass",
+      "all_values_and_dtypes": "pass",
+      "rows": 6250
+    },
+    {
+      "task": "d4_model_bias",
+      "public_download": "pass",
+      "sha256": "pass",
+      "all_values_and_dtypes": "pass",
+      "rows": 6250
+    },
+    {
+      "task": "d5_heavy_tail",
+      "public_download": "pass",
+      "sha256": "pass",
+      "all_values_and_dtypes": "pass",
+      "rows": 6250
+    },
+    {
+      "task": "d6_high_k",
+      "public_download": "pass",
+      "sha256": "pass",
+      "all_values_and_dtypes": "pass",
+      "rows": 17500
+    }
+  ]
+}

docs/croissant_validation/samson_unmixing.md

@@ -0,0 +1,293 @@
+# CROISSANT VALIDATION REPORT
+================================================================================
+## VALIDATION RESULTS
+--------------------------------------------------------------------------------
+Starting validation for file: samson_unmixing.json
+### JSON Format Validation
+✓
+The file is valid JSON.
+### Croissant Schema Validation
+✓
+The dataset passes Croissant validation.
+### Responsible AI Metadata
+✓
+All required Responsible AI metadata fields are present.
+### Records Generation Test
+✓
+Record set 'https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/samson_unmixing/samples' passed validation.
+## JSON-LD REFERENCE
+================================================================================
+```json
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/samson_unmixing/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#samson_unmixing",
+  "name": "SimplexTasks-12 \u2014 Samson hyperspectral unmixing",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/derived/samson_unmixing",
+  "description": "Samson hyperspectral unmixing is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 9025 samples with 3 simplex components. Frozen predictor: NMF abundance estimator. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/samson_unmixing.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "samson_unmixing"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "samson_unmixing.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/samson_unmixing.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "245140 B",
+      "sha256": "7be9f1cb90802b87866628afe8bab3d7c58e1671695e9497e9e1100dda7fa651"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "9025 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "The benchmark files are public and source-cited, but the upstream bundle did not include an explicit license file; downstream reuse should preserve attribution and the original benchmark context. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "One scene and three endmembers restrict geographic, material and sensor diversity; results may depend on spatial structure.",
+  "rai:personalSensitiveInformation": "The released arrays describe material abundances in a hyperspectral scene; the task documentation identifies no personal attributes in U and Y.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper\u2019s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": false,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://rslab.ut.ac.ir/data"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/samson_unmixing/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/samson_unmixing/Y.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-samson_unmixing",
+    "description": "Three-endmember abundance compositions paired with a frozen NMF abundance estimator. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. The public bundle did not ship an explicit license file; keep attribution with any downstream reuse. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/samson_unmixing.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Three-endmember abundance compositions paired with a frozen NMF abundance estimator. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. The public bundle did not ship an explicit license file; keep attribution with any downstream reuse. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/samson_unmixing.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}
+```

docs/croissant_validation/topics_20ng.md

@@ -0,0 +1,503 @@
+# CROISSANT VALIDATION REPORT
+================================================================================
+## VALIDATION RESULTS
+--------------------------------------------------------------------------------
+Starting validation for file: topics_20ng.json
+### JSON Format Validation
+✓
+The file is valid JSON.
+### Croissant Schema Validation
+✓
+The dataset passes Croissant validation.
+### Responsible AI Metadata
+✓
+All required Responsible AI metadata fields are present.
+### Records Generation Test
+✓
+Record set 'https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/topics_20ng/samples' passed validation.
+## JSON-LD REFERENCE
+================================================================================
+```json
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/topics_20ng/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#topics_20ng",
+  "name": "SimplexTasks-12 \u2014 20 Newsgroups topics",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/derived/topics_20newsgroups",
+  "description": "20 Newsgroups topics is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 7539 samples with 10 simplex components. Frozen predictor: TF-IDF to topic-mixture kNN regressor. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/topics_20newsgroups.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "topics_20ng"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "topics_20ng.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/topics_20ng.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "784025 B",
+      "sha256": "67832ac04d380db18c42ddde4cd56909d5d8644e975c57885078fcbdbe8cb604"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "7539 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_3",
+          "name": "U_3",
+          "description": "Frozen predictor output, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_4",
+          "name": "U_4",
+          "description": "Frozen predictor output, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_5",
+          "name": "U_5",
+          "description": "Frozen predictor output, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_6",
+          "name": "U_6",
+          "description": "Frozen predictor output, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_7",
+          "name": "U_7",
+          "description": "Frozen predictor output, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_8",
+          "name": "U_8",
+          "description": "Frozen predictor output, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_9",
+          "name": "U_9",
+          "description": "Frozen predictor output, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_9"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_3",
+          "name": "Y_3",
+          "description": "Benchmark target composition, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_4",
+          "name": "Y_4",
+          "description": "Benchmark target composition, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_5",
+          "name": "Y_5",
+          "description": "Benchmark target composition, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_6",
+          "name": "Y_6",
+          "description": "Benchmark target composition, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_7",
+          "name": "Y_7",
+          "description": "Benchmark target composition, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_8",
+          "name": "Y_8",
+          "description": "Benchmark target composition, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_9",
+          "name": "Y_9",
+          "description": "Benchmark target composition, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_9"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "Both the target and the predictor output are model-derived topic mixtures. The release exposes derived simplex arrays, not the raw text corpus. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "English newsgroup selection and model-generated topic targets can encode language, historical, topic and model biases.",
+  "rai:personalSensitiveInformation": "The source corpus can contain names, email addresses, political and religious discussions; the released U and Y are derived topic mixtures, without raw text. No formal privacy guarantee is claimed.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper\u2019s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": false,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "http://qwone.com/~jason/20Newsgroups/"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/topics_20newsgroups/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/topics_20newsgroups/Y.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-topics_20ng",
+    "description": "Fixed 10-topic compositions constructed once from the 20 Newsgroups corpus, with a TF-IDF-to-topic-mixture kNN regressor as the predictor. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. This release exposes only derived simplex arrays, not the raw text corpus. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/topics_20newsgroups.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "Fixed 10-topic compositions constructed once from the 20 Newsgroups corpus, with a TF-IDF-to-topic-mixture kNN regressor as the predictor. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. This release exposes only derived simplex arrays, not the raw text corpus. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/topics_20newsgroups.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}
+```

docs/croissant_validation/utkface_age_ldl.md

@@ -0,0 +1,503 @@
+# CROISSANT VALIDATION REPORT
+================================================================================
+## VALIDATION RESULTS
+--------------------------------------------------------------------------------
+Starting validation for file: utkface_age_ldl.json
+### JSON Format Validation
+✓
+The file is valid JSON.
+### Croissant Schema Validation
+✓
+The dataset passes Croissant validation.
+### Responsible AI Metadata
+✓
+All required Responsible AI metadata fields are present.
+### Records Generation Test
+✓
+Record set 'https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/utkface_age_ldl/samples' passed validation.
+## JSON-LD REFERENCE
+================================================================================
+```json
+{
+  "@context": {
+    "@language": "en",
+    "@vocab": "https://schema.org/",
+    "arrayShape": "cr:arrayShape",
+    "citeAs": "cr:citeAs",
+    "column": "cr:column",
+    "conformsTo": "dct:conformsTo",
+    "containedIn": "cr:containedIn",
+    "cr": "http://mlcommons.org/croissant/",
+    "data": {
+      "@id": "cr:data",
+      "@type": "@json"
+    },
+    "dataBiases": "cr:dataBiases",
+    "dataCollection": "cr:dataCollection",
+    "dataType": {
+      "@id": "cr:dataType",
+      "@type": "@vocab"
+    },
+    "dct": "http://purl.org/dc/terms/",
+    "extract": "cr:extract",
+    "field": "cr:field",
+    "fileProperty": "cr:fileProperty",
+    "fileObject": "cr:fileObject",
+    "fileSet": "cr:fileSet",
+    "format": "cr:format",
+    "includes": "cr:includes",
+    "isArray": "cr:isArray",
+    "isLiveDataset": "cr:isLiveDataset",
+    "jsonPath": "cr:jsonPath",
+    "key": "cr:key",
+    "md5": "cr:md5",
+    "parentField": "cr:parentField",
+    "path": "cr:path",
+    "personalSensitiveInformation": "cr:personalSensitiveInformation",
+    "recordSet": "cr:recordSet",
+    "references": "cr:references",
+    "regex": "cr:regex",
+    "repeated": "cr:repeated",
+    "replace": "cr:replace",
+    "sc": "https://schema.org/",
+    "separator": "cr:separator",
+    "source": "cr:source",
+    "subField": "cr:subField",
+    "transform": "cr:transform",
+    "rai": "http://mlcommons.org/croissant/RAI/",
+    "prov": "http://www.w3.org/ns/prov#",
+    "@base": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/croissant/utkface_age_ldl/",
+    "examples": {
+      "@id": "cr:examples",
+      "@type": "@json"
+    },
+    "samplingRate": "cr:samplingRate",
+    "equivalentProperty": "cr:equivalentProperty"
+  },
+  "@type": "sc:Dataset",
+  "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#utkface_age_ldl",
+  "name": "SimplexTasks-12 \u2014 UTKFace age label distributions",
+  "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/tree/main/data/derived/utkface_ldl",
+  "description": "UTKFace age label distributions is one task in SimplexTasks-12, the SimplexUQ evaluation collection. The public frozen release contains 23687 samples with 10 simplex components. Frozen predictor: Thumbnail feature PCA+kNN age-distribution regressor. This metadata describes a lossless Parquet table of the frozen NumPy arrays at public release revision ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501; sample_index preserves original row alignment. Task card: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/utkface_ldl.md. Supporting strata, configs and provenance remain in the same public repository.",
+  "conformsTo": "http://mlcommons.org/croissant/1.1",
+  "dct:conformsTo": "http://mlcommons.org/croissant/RAI/1.0",
+  "version": "0.1.0+parquet.ebf2d7be17db",
+  "datePublished": "2026-04-23",
+  "dateModified": "2026-10-16",
+  "creator": [
+    {
+      "@type": "sc:Person",
+      "name": "Liang You"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Hengyu Shi"
+    },
+    {
+      "@type": "sc:Person",
+      "name": "Dongwen Ou"
+    }
+  ],
+  "publisher": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "url": "https://huggingface.co/liangyou03"
+  },
+  "maintainer": {
+    "@type": "sc:Person",
+    "name": "Liang You",
+    "email": "liangyou03@pitt.edu"
+  },
+  "citeAs": "Liang You, Hengyu Shi, and Dongwen Ou. SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions. 2026.",
+  "license": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/LICENSE",
+  "sc:usageInfo": "Artifact license notice with task-specific source terms. The repository card and CITATION.cff use the other license category; there is no blanket CC-BY-4.0 grant. Format conversion does not change source rights or restrictions. See https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/ASSET_PROVENANCE_AND_TERMS.md and the linked task card.",
+  "isPartOf": {
+    "@type": "sc:Dataset",
+    "name": "SimplexTasks-12",
+    "url": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data"
+  },
+  "keywords": [
+    "SimplexUQ",
+    "conformal prediction",
+    "uncertainty evaluation",
+    "utkface_age_ldl"
+  ],
+  "distribution": [
+    {
+      "@type": "cr:FileObject",
+      "@id": "samples.parquet",
+      "name": "utkface_age_ldl.parquet",
+      "description": "Lossless tabular representation of frozen U/Y and, for synthetic tasks, R/sigma_true. Original precision, component order and sample row order retained.",
+      "contentUrl": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/data/parquet/utkface_age_ldl.parquet",
+      "encodingFormat": "application/x-parquet",
+      "contentSize": "925580 B",
+      "sha256": "e6b2f9a64d6a2da09e247093225d15b3ababd0387235c2285ce7af27f63f1a07"
+    }
+  ],
+  "recordSet": [
+    {
+      "@type": "cr:RecordSet",
+      "@id": "samples",
+      "name": "samples",
+      "description": "23687 samples from the frozen benchmark pool, one row per sample. This is not a prescribed train/calibration/test split; apply the published split policy.",
+      "key": {
+        "@id": "samples/sample_index"
+      },
+      "field": [
+        {
+          "@type": "cr:Field",
+          "@id": "samples/sample_index",
+          "name": "sample_index",
+          "description": "Zero-based original array row index; join key within this task.",
+          "dataType": "sc:Integer",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "sample_index"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_0",
+          "name": "U_0",
+          "description": "Frozen predictor output, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_1",
+          "name": "U_1",
+          "description": "Frozen predictor output, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_2",
+          "name": "U_2",
+          "description": "Frozen predictor output, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_3",
+          "name": "U_3",
+          "description": "Frozen predictor output, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_4",
+          "name": "U_4",
+          "description": "Frozen predictor output, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_5",
+          "name": "U_5",
+          "description": "Frozen predictor output, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_6",
+          "name": "U_6",
+          "description": "Frozen predictor output, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_7",
+          "name": "U_7",
+          "description": "Frozen predictor output, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_8",
+          "name": "U_8",
+          "description": "Frozen predictor output, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/U_9",
+          "name": "U_9",
+          "description": "Frozen predictor output, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "U_9"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_0",
+          "name": "Y_0",
+          "description": "Benchmark target composition, component 0 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_0"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_1",
+          "name": "Y_1",
+          "description": "Benchmark target composition, component 1 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_1"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_2",
+          "name": "Y_2",
+          "description": "Benchmark target composition, component 2 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_2"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_3",
+          "name": "Y_3",
+          "description": "Benchmark target composition, component 3 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_3"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_4",
+          "name": "Y_4",
+          "description": "Benchmark target composition, component 4 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_4"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_5",
+          "name": "Y_5",
+          "description": "Benchmark target composition, component 5 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_5"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_6",
+          "name": "Y_6",
+          "description": "Benchmark target composition, component 6 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_6"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_7",
+          "name": "Y_7",
+          "description": "Benchmark target composition, component 7 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_7"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_8",
+          "name": "Y_8",
+          "description": "Benchmark target composition, component 8 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_8"
+            }
+          }
+        },
+        {
+          "@type": "cr:Field",
+          "@id": "samples/Y_9",
+          "name": "Y_9",
+          "description": "Benchmark target composition, component 9 in original array order (zero-based).",
+          "dataType": "sc:Float",
+          "source": {
+            "fileObject": {
+              "@id": "samples.parquet"
+            },
+            "extract": {
+              "column": "Y_9"
+            }
+          }
+        }
+      ]
+    }
+  ],
+  "rai:dataLimitations": "The underlying face images are not redistributed and remain subject to the source dataset's research-use constraints. The benchmark ships derived features and simplex arrays only. Conclusions depend on the fixed predictor, score, evaluation groups and split protocol. Not established for production, clinical or demographic decision-making. Some historical summaries use protocols different from the final manuscript. Hosted metadata lists some auxiliary arrays that are not standalone public files; only the verified frozen arrays and their equivalent Parquet columns are represented here.",
+  "rai:dataBiases": "Face-image collection and age labels may encode demographic imbalance and annotation errors; group-wise demographic validity has not been established.",
+  "rai:personalSensitiveInformation": "Age-related distributions are derived from face imagery. The source includes age, gender and race annotations. This release\u2019s U and Y retain age-related information but exclude raw face images; no formal anonymisation guarantee is claimed.",
+  "rai:dataUseCases": "Intended to compare conformal uncertainty methods on fixed simplex targets and predictions, including marginal coverage, stratified coverage in prediction space, disparity, radius and runtime. Controlled simulations and the paper\u2019s task comparisons support this diagnostic purpose; they do not establish deployment safety, demographic fairness, clinical utility or suitability for training production models.",
+  "rai:dataSocialImpact": "May improve reproducibility and reveal uncertainty failures hidden by aggregate scores. Misuse could overstate reliability or turn task-specific comparisons into unsupported deployment decisions. Mitigations include task limitations, fixed-predictor evaluation, source attribution and exclusion of restricted raw assets.",
+  "rai:hasSyntheticData": false,
+  "prov:wasDerivedFrom": [
+    {
+      "@id": "https://susanqq.github.io/UTKFace/"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/utkface_ldl/U.npy"
+    },
+    {
+      "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/67f10fc96efab616fd327f47958e1cad8286786f/data/derived/utkface_ldl/Y.npy"
+    }
+  ],
+  "prov:wasGeneratedBy": {
+    "@type": "prov:Activity",
+    "@id": "https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data#processing-utkface_age_ldl",
+    "description": "A 10-bin age label distribution constructed from UTKFace ages and predicted from thumbnail image features using PCA+kNN regression. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. The package omits raw face images and keeps only derived features and simplex arrays. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/utkface_ldl.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json.",
+    "prov:wasAssociatedWith": [
+      {
+        "@type": "prov:Agent",
+        "name": "Liang You"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Hengyu Shi"
+      },
+      {
+        "@type": "prov:Agent",
+        "name": "Dongwen Ou"
+      }
+    ]
+  },
+  "rai:dataCollection": "A 10-bin age label distribution constructed from UTKFace ages and predicted from thumbnail image features using PCA+kNN regression. Derived task packaging and frozen predictions are by Liang You, Hengyu Shi and Dongwen Ou. The package omits raw face images and keeps only derived features and simplex arrays. Upstream collection and annotation are inherited from the source dataset; no new human annotation is documented for this release. Detailed upstream annotator demographics, agreement and collection geography are not reported in the hosted task card. Processing details and limitations: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/blob/67f10fc96efab616fd327f47958e1cad8286786f/task_cards/real/utkface_ldl.md. Rebuild code: https://github.com/liangyou03/SimplexUQ. On 2026-10-16 the released NumPy arrays were converted losslessly to Parquet, with sample_index and scalar component columns. Exact equality of values and floating-point dtypes was verified after reloading. No new labels, predictions or experiments were generated. Conversion evidence: https://huggingface.co/datasets/liangyou03/SimplexTasks-12-data/resolve/ebf2d7be17dbd2e8d21fb4eadd0af41d7e27d501/docs/parquet_verification.json."
+}
+```