dataset stringlengths 1 82 | model_name stringlengths 0 150 | paper_title stringlengths 22 175 | paper_date timestamp[ns]date 2023-05-02 00:00:00 2024-12-12 00:00:00 | paper_url stringlengths 32 35 | code_links listlengths 1 1 | prompts stringlengths 115 330 | answer stringlengths 1 22 | paper_text stringlengths 83 737k | year_bin stringclasses 2
values | benchmark_split stringclasses 1
value |
|---|---|---|---|---|---|---|---|---|---|---|
DUTS-TE | BiRefNet (DUTS, HRSOD, UHRSD) | Bilateral Reference for High-Resolution Dichotomous Image Segmentation | 2024-01-07T00:00:00 | https://arxiv.org/abs/2401.03407v6 | [
"https://github.com/zhengpeng7/birefnet"
] | In the paper 'Bilateral Reference for High-Resolution Dichotomous Image Segmentation', what MAE score did the BiRefNet (DUTS, HRSOD, UHRSD) model get on the DUTS-TE dataset
| 0.018 | Title: Bilateral Reference for High-Resolution Dichotomous Image Segmentation
Abstract: AbstractWe introduce a novel bilateral reference framework (BiRefNet) for high-resolution dichotomous image segmentation (DIS). It comprises two essential components: the localization module (LM) and the reconstruction module (RM) ... | 2024-2025 | public |
THUMOS’14 | RDFA-S6 (InternVideo2-6B) | Enhancing Temporal Action Localization: Advanced S6 Modeling with Recurrent Mechanism | 2024-07-18T00:00:00 | https://arxiv.org/abs/2407.13078v1 | [
"https://github.com/lsy0882/RDFA-S6"
] | In the paper 'Enhancing Temporal Action Localization: Advanced S6 Modeling with Recurrent Mechanism', what mAP IOU@0.5 score did the RDFA-S6 (InternVideo2-6B) model get on the THUMOS’14 dataset
| 78.2 | Title: Enhancing Temporal Action Localization:Advanced S6 Modeling with Recurrent Mechanism
Abstract: AbstractTemporal Action Localization (TAL) is a critical task in video analysis, identifying precise start and end times of actions. Existing methods like CNNs, RNNs, GCNs, and Transformers have limitations in capturi... | 2024-2025 | public |
fake | LightGBM + OpenAI embedding | PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning | 2024-03-31T00:00:00 | https://arxiv.org/abs/2404.00776v1 | [
"https://github.com/pyg-team/pytorch-frame"
] | In the paper 'PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning', what AUROC score did the LightGBM + OpenAI embedding model get on the fake dataset
| 0.966 | Title: PyTorch Frame: A Modular Framework for Multi-Modal Tabular Learning
Abstract: AbstractWe present PyTorch Frame, a PyTorch-based framework for deep learning over multi-modal tabular data. PyTorch Frame makes tabular deep learning easy by providing a PyTorch-based data structure to handle complex tabular data, in... | 2024-2025 | public |
MaleVis | Levit-MC | Accelerating Malware Classification: A Vision Transformer Solution | 2024-09-28T00:00:00 | https://arxiv.org/abs/2409.19461v1 | [
"https://github.com/Shrey-55/MalwareClassification"
] | In the paper 'Accelerating Malware Classification: A Vision Transformer Solution', what Accuracy score did the Levit-MC model get on the MaleVis dataset
| 96.6 | Title: Accelerating Malware Classification: A Vision Transformer Solution
Abstract: AbstractThe escalating frequency and scale of recent malware attacks underscore the urgent need for swift and precise malware classification in the ever-evolving cybersecurity landscape. Key challenges include accurately categorizing c... | 2024-2025 | public |
REDDIT-12K | G-Tuning | Fine-tuning Graph Neural Networks by Preserving Graph Generative Patterns | 2023-12-21T00:00:00 | https://arxiv.org/abs/2312.13583v1 | [
"https://github.com/zjunet/G-Tuning"
] | In the paper 'Fine-tuning Graph Neural Networks by Preserving Graph Generative Patterns', what Accuracy (10 fold) score did the G-Tuning model get on the REDDIT-12K dataset
| 42.80 | Title: Fine-tuning Graph Neural Networks by Preserving Graph Generative Patterns
Abstract: AbstractRecently, the paradigm of pre-training and fine-tuning graph neural networks has been intensively studied and applied in a wide range of graph mining tasks. Its success is generally attributed to the structural consisten... | 2023 | public |
IllusionVQA | GPT4-Vision 4-shot+CoT | IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models | 2024-03-23T00:00:00 | https://arxiv.org/abs/2403.15952v3 | [
"https://github.com/csebuetnlp/illusionvqa"
] | In the paper 'IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models', what Accuracy score did the GPT4-Vision 4-shot+CoT model get on the IllusionVQA dataset
| 49.7 | Title: IllusionVQA: A Challenging Optical Illusion Datasetfor Vision Language Models
Abstract: AbstractThe advent of Vision Language Models (VLM) has allowed researchers to investigate the visual understanding of a neural network using natural language. Beyond object classification and detection, VLMs are capable of v... | 2024-2025 | public |
Ant + Maze | STAR | Reconciling Spatial and Temporal Abstractions for Goal Representation | 2024-01-18T00:00:00 | https://arxiv.org/abs/2401.09870v2 | [
"https://github.com/cosynus-lix/STAR"
] | In the paper 'Reconciling Spatial and Temporal Abstractions for Goal Representation', what Return score did the STAR model get on the Ant + Maze dataset
| 0.85 | Title: Reconciling Spatial and Temporal Abstractions for Goal Representation
Abstract: AbstractGoal representation affects the performance of Hierarchical Reinforcement Learning (HRL) algorithms by decomposing the complex learning problem into easier subtasks. Recent studies show that representations that preserve tem... | 2024-2025 | public |
PASCAL-5i (1-Shot) | QCLNet (ResNet-101) | Quaternion-valued Correlation Learning for Few-Shot Semantic Segmentation | 2023-05-12T00:00:00 | https://arxiv.org/abs/2305.07283v3 | [
"https://github.com/zwzheng98/qclnet"
] | In the paper 'Quaternion-valued Correlation Learning for Few-Shot Semantic Segmentation', what Mean IoU score did the QCLNet (ResNet-101) model get on the PASCAL-5i (1-Shot) dataset
| 67 | Title: Quaternion-valued Correlation Learning for Few-Shot Semantic Segmentation
Abstract: AbstractFew-shot segmentation (FSS) aims to segment unseen classes given only a few annotated samples. Encouraging progress has been made for FSS by leveraging semantic features learned from base classes with sufficient training... | 2023 | public |
VeRi-776 | MBR4B-LAI (w/ RK) | Strength in Diversity: Multi-Branch Representation Learning for Vehicle Re-Identification | 2023-10-02T00:00:00 | https://arxiv.org/abs/2310.01129v1 | [
"https://github.com/videturfortuna/vehicle_reid_itsc2023"
] | In the paper 'Strength in Diversity: Multi-Branch Representation Learning for Vehicle Re-Identification', what mAP score did the MBR4B-LAI (w/ RK) model get on the VeRi-776 dataset
| 92.1 | Title: Strength in Diversity: Multi-Branch Representation Learning for Vehicle Re-Identification*
Abstract: AbstractThis paper presents an efficient and lightweight multi-branch deep architecture to improve vehicle re-identification (V-ReID). While most V-ReID work uses a combination of complex multi-branch architectu... | 2023 | public |
Weather2K850 (96) | MoLE-DLinear | Mixture-of-Linear-Experts for Long-term Time Series Forecasting | 2023-12-11T00:00:00 | https://arxiv.org/abs/2312.06786v3 | [
"https://github.com/rogerni/mole"
] | In the paper 'Mixture-of-Linear-Experts for Long-term Time Series Forecasting', what MSE score did the MoLE-DLinear model get on the Weather2K850 (96) dataset
| 0.474 | Title: Mixture-of-Linear-Experts for Long-term Time Series Forecasting
Abstract: AbstractLong-term time series forecasting (LTSF) aims to predict future values of a time series given the past values. The current state-of-the-art (SOTA) on this problem is attained in some cases bylinear-centricmodels, which primarily f... | 2023 | public |
AffectNet | FMAE | Representation Learning and Identity Adversarial Training for Facial Behavior Understanding | 2024-07-15T00:00:00 | https://arxiv.org/abs/2407.11243v1 | [
"https://github.com/forever208/fmae-iat"
] | In the paper 'Representation Learning and Identity Adversarial Training for Facial Behavior Understanding', what Accuracy (8 emotion) score did the FMAE model get on the AffectNet dataset
| 65.00 | Title: Representation Learning and Identity Adversarial Training for Facial Behavior Understanding
Abstract: AbstractFacial Action Unit (AU) detection has gained significant research attention as AUs contain complex expression information. In this paper, we unpack two fundamental factors in AU detection: data and subj... | 2024-2025 | public |
Watercolor2k | CDDMSL | Semi-Supervised Domain Generalization for Object Detection via Language-Guided Feature Alignment | 2023-09-24T00:00:00 | https://arxiv.org/abs/2309.13525v1 | [
"https://github.com/sinamalakouti/CDDMSL"
] | In the paper 'Semi-Supervised Domain Generalization for Object Detection via Language-Guided Feature Alignment', what MAP score did the CDDMSL model get on the Watercolor2k dataset
| 49.8 | Title: Semi-Supervised Domain Generalization for Object Detection via Language-Guided Feature Alignment
Abstract: AbstractExisting domain adaptation (DA) and generalization (DG) methods in object detection enforce feature alignment in the visual space but face challenges like object appearance variability and scene co... | 2023 | public |
Leaderboard | LLaVA-Plus (13B) | LLaVA-Plus: Learning to Use Tools for Creating Multimodal Agents | 2023-11-09T00:00:00 | https://arxiv.org/abs/2311.05437v1 | [
"https://github.com/LLaVA-VL/LLaVA-Plus-Codebase"
] | In the paper 'LLaVA-Plus: Learning to Use Tools for Creating Multimodal Agents', what ELO Rating score did the LLaVA-Plus (13B) model get on the Leaderboard dataset
| 1203 | Title: LLaVA-Plus: Learning to Use Tools forCreating Multimodal Agents
Abstract: AbstractThis paper presents LLaVA-Plus (LargeLanguageandVisionAssistants thatPlug andLearn toUseSkills), a general-purpose multimodal assistant trained using an end-to-end approach that systematically expands the capabilities of large mul... | 2023 | public |
ETTh2 (192) Multivariate | RLinear | Revisiting Long-term Time Series Forecasting: An Investigation on Linear Mapping | 2023-05-18T00:00:00 | https://arxiv.org/abs/2305.10721v1 | [
"https://github.com/plumprc/rtsf"
] | In the paper 'Revisiting Long-term Time Series Forecasting: An Investigation on Linear Mapping', what MSE score did the RLinear model get on the ETTh2 (192) Multivariate dataset
| 0.319 | Title: Revisiting Long-term Time Series Forecasting:An Investigation on Linear Mapping
Abstract: AbstractLong-term time series forecasting has gained significant attention in recent years. While there are various specialized designs for capturing temporal dependency, previous studies have demonstrated that a single li... | 2023 | public |
ImageNet 512x512 | DiffiT | DiffiT: Diffusion Vision Transformers for Image Generation | 2023-12-04T00:00:00 | https://arxiv.org/abs/2312.02139v3 | [
"https://github.com/nvlabs/diffit"
] | In the paper 'DiffiT: Diffusion Vision Transformers for Image Generation', what FID score did the DiffiT model get on the ImageNet 512x512 dataset
| 2.67 | Title: DiffiT:Diffusion VisionTransformers for Image Generation
Abstract: AbstractDiffusion models with their powerful expressivity and high sample quality have enabled many new applications and use-cases in various domains. For sample generation, these models rely on a denoising neural network that generates images b... | 2023 | public |
DeLiVER | StitchFusion (RGB-D-Event) | StitchFusion: Weaving Any Visual Modalities to Enhance Multimodal Semantic Segmentation | 2024-08-02T00:00:00 | https://arxiv.org/abs/2408.01343v1 | [
"https://github.com/libingyu01/stitchfusion-stitchfusion-weaving-any-visual-modalities-to-enhance-multimodal-semantic-segmentation"
] | In the paper 'StitchFusion: Weaving Any Visual Modalities to Enhance Multimodal Semantic Segmentation', what mIoU score did the StitchFusion (RGB-D-Event) model get on the DeLiVER dataset
| 66.03 | Title: StitchFusion: Weaving Any Visual Modalities to Enhance Multimodal Semantic Segmentation
Abstract: AbstractMultimodal semantic segmentation shows significant potential for enhancing segmentation accuracy in complex scenes. However, current methods often incorporate specialized feature fusion modules tailored to ... | 2024-2025 | public |
CROHME 2019 | TAMER | TAMER: Tree-Aware Transformer for Handwritten Mathematical Expression Recognition | 2024-08-16T00:00:00 | https://arxiv.org/abs/2408.08578v2 | [
"https://github.com/qingzhenduyu/tamer"
] | In the paper 'TAMER: Tree-Aware Transformer for Handwritten Mathematical Expression Recognition', what ExpRate score did the TAMER model get on the CROHME 2019 dataset
| 61.97 | Title: TAMER: Tree-Aware Transformer for Handwritten Mathematical Expression Recognition
Abstract: AbstractHandwritten Mathematical Expression Recognition (HMER) has extensive applications in automated grading and office automation. However, existing sequence-based decoding methods, which directly predictLaTeXsequence... | 2024-2025 | public |
ETTm2 (192) Multivariate | TSMixer | TSMixer: Lightweight MLP-Mixer Model for Multivariate Time Series Forecasting | 2023-06-14T00:00:00 | https://arxiv.org/abs/2306.09364v4 | [
"https://github.com/ibm/tsfm"
] | In the paper 'TSMixer: Lightweight MLP-Mixer Model for Multivariate Time Series Forecasting', what MSE score did the TSMixer model get on the ETTm2 (192) Multivariate dataset
| 0.219 | Title: TSMixer: Lightweight MLP-Mixer Model forMultivariate Time Series Forecasting
Abstract: Abstract.Transformers have gained popularity in time series forecasting for their ability to capture long-sequence interactions. However, their memory and compute-intensive requirements pose a critical bottleneck for long-ter... | 2023 | public |
LAMBADA | PaLM 2-L (one-shot) | PaLM 2 Technical Report | 2023-05-17T00:00:00 | https://arxiv.org/abs/2305.10403v3 | [
"https://github.com/eternityyw/tram-benchmark"
] | In the paper 'PaLM 2 Technical Report', what Accuracy score did the PaLM 2-L (one-shot) model get on the LAMBADA dataset
| 86.9 | Title: PaLM 2 Technical Report
Abstract: AbstractWe introduce PaLM 2, a new state-of-the-art language model that has better multilingual
and reasoning capabilities and is more compute-efficient than its
predecessor PaLM. PaLM 2 is a Transformer-based model trained using a mixture of objectives. Through extensive evalu... | 2023 | public |
CUHK-PEDES | TBPS-CLIP (ViT-B/16) | An Empirical Study of CLIP for Text-based Person Search | 2023-08-19T00:00:00 | https://arxiv.org/abs/2308.10045v2 | [
"https://github.com/flame-chasers/tbps-clip"
] | In the paper 'An Empirical Study of CLIP for Text-based Person Search', what R@1 score did the TBPS-CLIP (ViT-B/16) model get on the CUHK-PEDES dataset
| 73.54 | Title: An Empirical Study of CLIP for Text-based Person Search
Abstract: AbstractText-based Person Search (TBPS) aims to retrieve the person images using natural language descriptions. Recently, Contrastive Language Image Pretraining (CLIP), a universal large cross-modal vision-language pre-training model, has remarka... | 2023 | public |
EconLogicQA | Mistral-7B-v0.2 | EconLogicQA: A Question-Answering Benchmark for Evaluating Large Language Models in Economic Sequential Reasoning | 2024-05-13T00:00:00 | https://arxiv.org/abs/2405.07938v2 | [
"https://github.com/yinzhu-quan/lm-evaluation-harness"
] | In the paper 'EconLogicQA: A Question-Answering Benchmark for Evaluating Large Language Models in Economic Sequential Reasoning', what Accuracy score did the Mistral-7B-v0.2 model get on the EconLogicQA dataset
| 0.2615 | Title: EconLogicQA: A Question-Answering Benchmark for EvaluatingLarge Language Models in Economic Sequential Reasoning
Abstract: AbstractIn this paper, we introduce EconLogicQA, a rigorous benchmark designed to assess the sequential reasoning capabilities of large language models (LLMs) within the intricate realms of... | 2024-2025 | public |
UBnormal | MoCoDAD | Multimodal Motion Conditioned Diffusion Model for Skeleton-based Video Anomaly Detection | 2023-07-14T00:00:00 | https://arxiv.org/abs/2307.07205v3 | [
"https://github.com/aleflabo/MoCoDAD"
] | In the paper 'Multimodal Motion Conditioned Diffusion Model for Skeleton-based Video Anomaly Detection', what AUC score did the MoCoDAD model get on the UBnormal dataset
| 68.3% | Title: Multimodal Motion Conditioned Diffusion Model for Skeleton-based Video Anomaly Detection
Abstract: AbstractAnomalies are rare and anomaly detection is often therefore framed as One-Class Classification (OCC), i.e. trained solely on normalcy. Leading OCC techniques constrain the latent representations of normal1... | 2023 | public |
MM-Vet | ShareGPT4V-7B | ShareGPT4V: Improving Large Multi-Modal Models with Better Captions | 2023-11-21T00:00:00 | https://arxiv.org/abs/2311.12793v2 | [
"https://github.com/InternLM/InternLM-XComposer/tree/main/projects/ShareGPT4V"
] | In the paper 'ShareGPT4V: Improving Large Multi-Modal Models with Better Captions', what GPT-4 score score did the ShareGPT4V-7B model get on the MM-Vet dataset
| 37.6 | Title: ShareGPT4V: Improving Large Multi-Modal Models with Better Captions
Abstract: AbstractIn the realm of large multi-modal models (LMMs), efficient modality alignment is crucial yet often constrained by the scarcity of high-quality image-text data. To address this bottleneck, we introduce the ShareGPT4V dataset, a... | 2023 | public |
LTCC | FIRe2 | Exploring Fine-Grained Representation and Recomposition for Cloth-Changing Person Re-Identification | 2023-08-21T00:00:00 | https://arxiv.org/abs/2308.10692v2 | [
"https://github.com/qizaowang/fire-ccreid"
] | In the paper 'Exploring Fine-Grained Representation and Recomposition for Cloth-Changing Person Re-Identification', what Rank-1 score did the FIRe2 model get on the LTCC dataset
| 44.6 | Title: Exploring Fine-Grained Representation and Recompositionfor Cloth-Changing Person Re-Identification
Abstract: AbstractCloth-changing person Re-IDentification (Re-ID) is a particularly challenging task, suffering from two limitations of inferior identity-relevant features and limited training samples. Existing me... | 2023 | public |
RefCoCo val | MagNet | Mask Grounding for Referring Image Segmentation | 2023-12-19T00:00:00 | https://arxiv.org/abs/2312.12198v2 | [
"https://github.com/yxchng/mask-grounding"
] | In the paper 'Mask Grounding for Referring Image Segmentation', what Overall IoU score did the MagNet model get on the RefCoCo val dataset
| 75.24 | Title: Mask Grounding for Referring Image Segmentation
Abstract: AbstractReferring Image Segmentation (RIS) is a challenging task that requires an algorithm to segment objects referred by free-form language expressions. Despite significant progress in recent years, most state-of-the-art (SOTA) methods still suffer fro... | 2023 | public |
TAO | AED (Co-DETR) | Associate Everything Detected: Facilitating Tracking-by-Detection to the Unknown | 2024-09-14T00:00:00 | https://arxiv.org/abs/2409.09293v1 | [
"https://github.com/balabooooo/aed"
] | In the paper 'Associate Everything Detected: Facilitating Tracking-by-Detection to the Unknown', what TETA score did the AED (Co-DETR) model get on the TAO dataset
| 55.3 | Title: Associate Everything Detected: Facilitating Tracking-by-Detection to the Unknown
Abstract: AbstractMulti-object tracking (MOT) emerges as a pivotal and highly promising branch in the field of computer vision. Classical closed-vocabulary MOT (CV-MOT) methods aim to track objects of predefined categories. Recentl... | 2024-2025 | public |
CausalGym | Difference-in-means | CausalGym: Benchmarking causal interpretability methods on linguistic tasks | 2024-02-19T00:00:00 | https://arxiv.org/abs/2402.12560v1 | [
"https://github.com/aryamanarora/causalgym"
] | In the paper 'CausalGym: Benchmarking causal interpretability methods on linguistic tasks', what Log odds-ratio (pythia-6.9b) score did the Difference-in-means model get on the CausalGym dataset
| 2.91 | Title: CausalGym: Benchmarking causal interpretability methodson linguistic tasks
Abstract: AbstractLanguage models (LMs) have proven to be powerful tools for psycholinguistic research, but most prior work has focused on purely behavioural measures (e.g., surprisal comparisons). At the same time, research in model int... | 2024-2025 | public |
CoNLL-2014 Shared Task | Majority-voting ensemble on best 7 models | Pillars of Grammatical Error Correction: Comprehensive Inspection Of Contemporary Approaches In The Era of Large Language Models | 2024-04-23T00:00:00 | https://arxiv.org/abs/2404.14914v1 | [
"https://github.com/grammarly/pillars-of-gec"
] | In the paper 'Pillars of Grammatical Error Correction: Comprehensive Inspection Of Contemporary Approaches In The Era of Large Language Models', what F0.5 score did the Majority-voting ensemble on best 7 models model get on the CoNLL-2014 Shared Task dataset
| 71.8 | Title: Pillars of Grammatical Error Correction: Comprehensive Inspection Of Contemporary Approaches In The Era of Large Language Models
Abstract: AbstractIn this paper, we carry out experimental research on Grammatical Error Correction, delving into the nuances of single-model systems, comparing the efficiency of ense... | 2024-2025 | public |
WHU-CD | CGNet | Change Guiding Network: Incorporating Change Prior to Guide Change Detection in Remote Sensing Imagery | 2024-04-14T00:00:00 | https://arxiv.org/abs/2404.09179v1 | [
"https://github.com/chengxihan/cgnet-cd"
] | In the paper 'Change Guiding Network: Incorporating Change Prior to Guide Change Detection in Remote Sensing Imagery', what F1 score did the CGNet model get on the WHU-CD dataset
| 92.59 | Warning: Very short content extracted (0 chars). URL may not be properly formatted. | 2024-2025 | public |
V2XSet | V2X-AHD | V2X-AHD:Vehicle-to-Everything Cooperation Perception via Asymmetric Heterogenous Distillation Network | 2023-10-10T00:00:00 | https://arxiv.org/abs/2310.06603v1 | [
"https://github.com/feeling0414-lab/V2X-AHD"
] | In the paper 'V2X-AHD:Vehicle-to-Everything Cooperation Perception via Asymmetric Heterogenous Distillation Network', what AP0.5 (Perfect) score did the V2X-AHD model get on the V2XSet dataset
| 0.855 | Title: V2X-AHD:Vehicle-to-Everything Cooperation Perception via Asymmetric Heterogenous Distillation Network
Abstract: AbstractObject detection is the central issue of intelligent traffic systems, and recent advancements in single-vehicle lidar-based 3D detection indicate that it can provide accurate position informat... | 2023 | public |
MATH | OpenMath-Mistral-7B (w/ code, SC, k=50) | OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset | 2024-02-15T00:00:00 | https://arxiv.org/abs/2402.10176v2 | [
"https://github.com/kipok/nemo-skills"
] | In the paper 'OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset', what Accuracy score did the OpenMath-Mistral-7B (w/ code, SC, k=50) model get on the MATH dataset
| 57.2 | Title: OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset
Abstract: AbstractRecent work has shown the immense potential of synthetically generated datasets for training large language models (LLMs), especially for acquiring targeted skills. Current large-scale math instruction tuning datasets such as Me... | 2024-2025 | public |
ImageNet-100 (Class-IL, 5T) | MoCo + CaSSLe | Regularizing with Pseudo-Negatives for Continual Self-Supervised Learning | 2023-06-08T00:00:00 | https://arxiv.org/abs/2306.05101v2 | [
"https://github.com/csm9493/PNR"
] | In the paper 'Regularizing with Pseudo-Negatives for Continual Self-Supervised Learning', what Top 1 Accuracy score did the MoCo + CaSSLe model get on the ImageNet-100 (Class-IL, 5T) dataset
| 63.49 | Title: Sy-CON: Symmetric Contrastive Loss forContinual Self-Supervised Representation Learning
Abstract: AbstractWe introduce a novel and general loss function, called Symmetric Contrastive (Sy-CON) loss, for effective continual self-supervised learning (CSSL). We first argue that the conventional loss form of continu... | 2023 | public |
miniF2F-test | MMOS-DeepSeekMath-7B | An Empirical Study of Data Ability Boundary in LLMs' Math Reasoning | 2024-02-23T00:00:00 | https://arxiv.org/abs/2403.00799v1 | [
"https://github.com/cyzhh/MMOS"
] | In the paper 'An Empirical Study of Data Ability Boundary in LLMs' Math Reasoning', what Pass@1 score did the MMOS-DeepSeekMath-7B model get on the miniF2F-test dataset
| 28.3 | Title: An Empirical Study of Data Ability Boundary in LLMs’ Math Reasoning
Abstract: AbstractLarge language models (LLMs) are displaying emergent abilities for math reasoning tasks,
and there is a growing attention on enhancing the ability of open-source LLMs through supervised fine-tuning (SFT). In this paper, we aim... | 2024-2025 | public |
CIRR | SPN4CIR | Improving Composed Image Retrieval via Contrastive Learning with Scaling Positives and Negatives | 2024-04-17T00:00:00 | https://arxiv.org/abs/2404.11317v2 | [
"https://github.com/BUAADreamer/SPN4CIR"
] | In the paper 'Improving Composed Image Retrieval via Contrastive Learning with Scaling Positives and Negatives', what (Recall@5+Recall_subset@1)/2 score did the SPN4CIR model get on the CIRR dataset
| 82.69 | Title: Improving Composed Image Retrieval via Contrastive Learning with Scaling Positives and Negatives
Abstract: AbstractThe Composed Image Retrieval (CIR) task aims to retrieve target images using a composed query consisting of a reference image and a modified text. Advanced methods often utilize contrastive learnin... | 2024-2025 | public |
AFAD | ResNet-50-DLDL | A Call to Reflect on Evaluation Practices for Age Estimation: Comparative Analysis of the State-of-the-Art and a Unified Benchmark | 2023-07-10T00:00:00 | https://arxiv.org/abs/2307.04570v3 | [
"https://github.com/paplhjak/facial-age-estimation-benchmark"
] | In the paper 'A Call to Reflect on Evaluation Practices for Age Estimation: Comparative Analysis of the State-of-the-Art and a Unified Benchmark', what MAE score did the ResNet-50-DLDL model get on the AFAD dataset
| 3.14 | Title: A Call to Reflect on Evaluation Practices for Age Estimation:Comparative Analysis of the State-of-the-Art and a Unified Benchmark
Abstract: AbstractComparing different age estimation methods poses a challenge due to the unreliability of published results stemming from inconsistencies in the benchmarking process... | 2023 | public |
CBSD68 sigm75 | MeD | Multi-view Self-supervised Disentanglement for General Image Denoising | 2023-09-10T00:00:00 | https://arxiv.org/abs/2309.05049v1 | [
"https://github.com/chqwer2/multi-view-self-supervised-disentanglement-denoising"
] | In the paper 'Multi-view Self-supervised Disentanglement for General Image Denoising', what PSNR/SSIM score did the MeD model get on the CBSD68 sigm75 dataset
| 25.40/0.6645 | Title: Multi-view Self-supervised Disentanglement for General Image Denoising
Abstract: AbstractWith its significant performance improvements, the deep learning paradigm has become a standard tool for modern image denoisers. While promising performance has been shown on seen noise distributions, existing approaches of... | 2023 | public |
EMNIST-Letters | Spiking-Diffusion | Spiking-Diffusion: Vector Quantized Discrete Diffusion Model with Spiking Neural Networks | 2023-08-20T00:00:00 | https://arxiv.org/abs/2308.10187v4 | [
"https://github.com/Arktis2022/Spiking-Diffusion"
] | In the paper 'Spiking-Diffusion: Vector Quantized Discrete Diffusion Model with Spiking Neural Networks', what FID score did the Spiking-Diffusion model get on the EMNIST-Letters dataset
| 67.41 | Title: Spiking-Diffusion: Vector Quantized Discrete Diffusion Model with Spiking Neural Networks
Abstract: AbstractSpiking neural networks (SNNs) have tremendous potential for energy-efficient neuromorphic chips due to their binary and event-driven architecture. SNNs have been primarily used in classification tasks, b... | 2023 | public |
COCO | CM3Leon-7B | Scaling Autoregressive Multi-Modal Models: Pretraining and Instruction Tuning | 2023-09-05T00:00:00 | https://arxiv.org/abs/2309.02591v1 | [
"https://github.com/kyegomez/CM3Leon"
] | In the paper 'Scaling Autoregressive Multi-Modal Models: Pretraining and Instruction Tuning', what FID score did the CM3Leon-7B model get on the COCO dataset
| 4.88 | Title: Scaling Autoregressive Multi-Modal Models: Pretraining and Instruction Tuning
Abstract: AbstractWe present CM3Leon (pronounced “Chameleon”), a retrieval-augmented, token-based, decoder-only multi-modal language model capable of generating and infilling both text and images. CM3Leon uses the CM3 multi-modal arch... | 2023 | public |
AudioSet | DASS-Medium (Audio-only, single) | DASS: Distilled Audio State Space Models Are Stronger and More Duration-Scalable Learners | 2024-07-04T00:00:00 | https://arxiv.org/abs/2407.04082v1 | [
"https://github.com/Saurabhbhati/DASS"
] | In the paper 'DASS: Distilled Audio State Space Models Are Stronger and More Duration-Scalable Learners', what Test mAP score did the DASS-Medium (Audio-only, single) model get on the AudioSet dataset
| 0.476 | Title: DASS: Distilled Audio State Space ModelsAre Stronger and More Duration-Scalable Learners
Abstract: AbstractState-space models (SSMs) have emerged as an alternative to Transformers for audio modeling due to their high computational efficiency with long inputs. While recent efforts on Audio SSMs have reported enc... | 2024-2025 | public |
EconLogicQA | Llama-2-7B | EconLogicQA: A Question-Answering Benchmark for Evaluating Large Language Models in Economic Sequential Reasoning | 2024-05-13T00:00:00 | https://arxiv.org/abs/2405.07938v2 | [
"https://github.com/yinzhu-quan/lm-evaluation-harness"
] | In the paper 'EconLogicQA: A Question-Answering Benchmark for Evaluating Large Language Models in Economic Sequential Reasoning', what Accuracy score did the Llama-2-7B model get on the EconLogicQA dataset
| 0.0077 | Title: EconLogicQA: A Question-Answering Benchmark for EvaluatingLarge Language Models in Economic Sequential Reasoning
Abstract: AbstractIn this paper, we introduce EconLogicQA, a rigorous benchmark designed to assess the sequential reasoning capabilities of large language models (LLMs) within the intricate realms of... | 2024-2025 | public |
IndustReal | MViT-V2 | IndustReal: A Dataset for Procedure Step Recognition Handling Execution Errors in Egocentric Videos in an Industrial-Like Setting | 2023-10-26T00:00:00 | https://arxiv.org/abs/2310.17323v1 | [
"https://github.com/timschoonbeek/industreal"
] | In the paper 'IndustReal: A Dataset for Procedure Step Recognition Handling Execution Errors in Egocentric Videos in an Industrial-Like Setting', what Top-1 score did the MViT-V2 model get on the IndustReal dataset
| 65.25 | Title: IndustReal: A Dataset for Procedure Step Recognition HandlingExecution Errors in Egocentric Videos in an Industrial-Like Setting
Abstract: AbstractAlthough action recognition for procedural tasks has received notable attention, it has a fundamental flaw in that no measure of success for actions is provided. Thi... | 2023 | public |
MVTec LOCO AD | SINBAD+EfficientAD | Set Features for Anomaly Detection | 2023-11-24T00:00:00 | https://arxiv.org/abs/2311.14773v3 | [
"https://github.com/NivC/SINBAD"
] | In the paper 'Set Features for Anomaly Detection', what Avg. Detection AUROC score did the SINBAD+EfficientAD model get on the MVTec LOCO AD dataset
| 94.2 | Title: Set Features for Anomaly Detection
Abstract: AbstractThis paper proposes set features for detecting anomalies in samples that consist of unusual combinations of normal elements. Many leading methods
discover anomalies by detecting an unusual part of a sample. For example, state-of-the-art segmentation-based app... | 2023 | public |
Wiki-CS | GraphSAGE | Half-Hop: A graph upsampling approach for slowing down message passing | 2023-08-17T00:00:00 | https://arxiv.org/abs/2308.09198v1 | [
"https://github.com/nerdslab/halfhop"
] | In the paper 'Half-Hop: A graph upsampling approach for slowing down message passing', what Accuracy score did the GraphSAGE model get on the Wiki-CS dataset
| 83.67 | Title: Half-Hop: A graph upsampling approach for slowing down message passing
Abstract: AbstractMessage passing neural networks have shown a lot of success on graph-structured data. However, there are many instances where message passing can lead to over-smoothing or fail when neighboring nodes belong to different cla... | 2023 | public |
NYU Depth v2 | SMMCL (ResNet-101) | Understanding Dark Scenes by Contrasting Multi-Modal Observations | 2023-08-23T00:00:00 | https://arxiv.org/abs/2308.12320v2 | [
"https://github.com/palmdong/smmcl"
] | In the paper 'Understanding Dark Scenes by Contrasting Multi-Modal Observations', what Mean IoU score did the SMMCL (ResNet-101) model get on the NYU Depth v2 dataset
| 52.5% | Title: Understanding Dark Scenes by Contrasting Multi-Modal Observations
Abstract: AbstractUnderstanding dark scenes based on multi-modal image data is challenging, as both the visible and auxiliary modalities provide limited semantic information for the task. Previous methods focus on fusing the two modalities but ne... | 2023 | public |
spider | T5-3B+NatSQL+Token Preprocessing | Improving Generalization in Language Model-Based Text-to-SQL Semantic Parsing: Two Simple Semantic Boundary-Based Techniques | 2023-05-27T00:00:00 | https://arxiv.org/abs/2305.17378v1 | [
"https://github.com/dakingrai/ood-generalization-semantic-boundary-techniques"
] | In the paper 'Improving Generalization in Language Model-Based Text-to-SQL Semantic Parsing: Two Simple Semantic Boundary-Based Techniques', what Exact Match Accuracy (Dev) score did the T5-3B+NatSQL+Token Preprocessing model get on the spider dataset
| 69.4 | Title: Improving Generalization in Language Model-Based Text-to-SQL Semantic Parsing: Two Simple Semantic Boundary-Based Techniques
Abstract: AbstractCompositional and domain generalization present significant challenges in semantic parsing, even for state-of-the-art semantic parsers based on pre-trained language mode... | 2023 | public |
Office-Home | PGA (ViT-B/16) | Enhancing Domain Adaptation through Prompt Gradient Alignment | 2024-06-13T00:00:00 | https://arxiv.org/abs/2406.09353v2 | [
"https://github.com/viethoang1512/pga"
] | In the paper 'Enhancing Domain Adaptation through Prompt Gradient Alignment', what Accuracy score did the PGA (ViT-B/16) model get on the Office-Home dataset
| 85.1 | Abstract: AbstractPrior Unsupervised Domain Adaptation (UDA) methods often aim to train a domain-invariant feature extractor, which may hinder the model from learning sufficiently discriminative features. To tackle this, a line of works based on prompt learning leverages the power of large-scale pre-trained vision-lang... | 2024-2025 | public |
ARMBench | RISE (R101) | Robot Instance Segmentation with Few Annotations for Grasping | 2024-07-01T00:00:00 | https://arxiv.org/abs/2407.01302v1 | [
"https://github.com/mkimhi/RISE"
] | In the paper 'Robot Instance Segmentation with Few Annotations for Grasping', what AP50 score did the RISE (R101) model get on the ARMBench dataset
| 84.74 | Title: Robot Instance Segmentation with Few Annotations for Grasping
Abstract: AbstractThe ability of robots to manipulate objects relies heavily on their aptitude for visual perception. In domains characterized by cluttered scenes and high object variability, most methods call for vast labeled datasets, laboriously h... | 2024-2025 | public |
CIFAR-10 (250 Labels, ImageNet-100 Unlabeled) | UnMixMatch | Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data | 2023-06-02T00:00:00 | https://arxiv.org/abs/2306.01222v2 | [
"https://github.com/shuvenduroy/unmixmatch"
] | In the paper 'Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data', what Accuracy score did the UnMixMatch model get on the CIFAR-10 (250 Labels, ImageNet-100 Unlabeled) dataset
| 68.72 | Title: Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data
Abstract: AbstractWe propose UnMixMatch, a semi-supervised learning framework which can learn effective representations from unconstrained unlabelled data in order to scale up performance. Most existing semi-supervised methods rely on the as... | 2023 | public |
ImageNet 256x256 | ACDiT | ACDiT: Interpolating Autoregressive Conditional Modeling and Diffusion Transformer | 2024-12-10T00:00:00 | https://arxiv.org/abs/2412.07720v1 | [
"https://github.com/thunlp/acdit"
] | In the paper 'ACDiT: Interpolating Autoregressive Conditional Modeling and Diffusion Transformer', what FID score did the ACDiT model get on the ImageNet 256x256 dataset
| 2.37 | Title: ACDiT: Interpolating Autoregressive Conditional Modeling andDiffusion Transformer
Abstract: AbstractThe recent surge of interest in comprehensive multimodal models has necessitated the unification of diverse modalities. However, the unification suffers from disparate methodologies. Continuous visual generation ... | 2024-2025 | public |
BigEarthNet-S1 (official test set) | ViT-S/16 | Feature Guided Masked Autoencoder for Self-supervised Learning in Remote Sensing | 2023-10-28T00:00:00 | https://arxiv.org/abs/2310.18653v1 | [
"https://github.com/zhu-xlab/fgmae"
] | In the paper 'Feature Guided Masked Autoencoder for Self-supervised Learning in Remote Sensing', what mAP (micro) score did the ViT-S/16 model get on the BigEarthNet-S1 (official test set) dataset
| 79.5 | Title: Feature Guided Masked Autoencoder for Self-supervised Learning in Remote Sensing
Abstract: AbstractSelf-supervised learning guided by masked image modelling, such as Masked AutoEncoder (MAE), has attracted wide attention for pretraining vision transformers in remote sensing. However, MAE tends to excessively fo... | 2023 | public |
KIT Motion-Language | DiverseMotion | DiverseMotion: Towards Diverse Human Motion Generation via Discrete Diffusion | 2023-09-04T00:00:00 | https://arxiv.org/abs/2309.01372v1 | [
"https://github.com/axdfhj/mdd"
] | In the paper 'DiverseMotion: Towards Diverse Human Motion Generation via Discrete Diffusion', what FID score did the DiverseMotion model get on the KIT Motion-Language dataset
| 0.468 | Title: DiverseMotion: Towards Diverse Human Motion Generation viaDiscrete Diffusion
Abstract: AbstractWe present DiverseMotion, a new approach for synthesizing high-quality human motions conditioned on textual descriptions while preserving motion diversity. Despite the recent significant process in text-based human mo... | 2023 | public |
OVIS validation | DVIS++(VIT-L,Offline) | DVIS++: Improved Decoupled Framework for Universal Video Segmentation | 2023-12-20T00:00:00 | https://arxiv.org/abs/2312.13305v1 | [
"https://github.com/zhang-tao-whu/DVIS_Plus"
] | In the paper 'DVIS++: Improved Decoupled Framework for Universal Video Segmentation', what mask AP score did the DVIS++(VIT-L,Offline) model get on the OVIS validation dataset
| 53.4 | Title: DVIS++: Improved Decoupled Framework for Universal Video Segmentation
Abstract: AbstractWe present theDecoupledVIdeoSegmentation (DVIS) framework, a novel approach for the challenging task of universal video segmentation, including video instance segmentation (VIS), video semantic segmentation (VSS), and video ... | 2023 | public |
HistGen WSI-Report Dataset | HistGen | HistGen: Histopathology Report Generation via Local-Global Feature Encoding and Cross-modal Context Interaction | 2024-03-08T00:00:00 | https://arxiv.org/abs/2403.05396v2 | [
"https://github.com/dddavid4real/HistGen"
] | In the paper 'HistGen: Histopathology Report Generation via Local-Global Feature Encoding and Cross-modal Context Interaction', what BLEU-4 score did the HistGen model get on the HistGen WSI-Report Dataset dataset
| 0.184 | Title: HistGen: Histopathology Report Generation via Local-Global Feature Encoding and Cross-modal Context Interaction
Abstract: AbstractHistopathology serves as the gold standard in cancer diagnosis, with clinical reports being vital in interpreting and understanding this process, guiding cancer treatment and patient... | 2024-2025 | public |
ModelNet40 | Point-JEPA (voting) | Point-JEPA: A Joint Embedding Predictive Architecture for Self-Supervised Learning on Point Cloud | 2024-04-25T00:00:00 | https://arxiv.org/abs/2404.16432v4 | [
"https://github.com/Ayumu-J-S/Point-JEPA"
] | In the paper 'Point-JEPA: A Joint Embedding Predictive Architecture for Self-Supervised Learning on Point Cloud', what Overall Accuracy score did the Point-JEPA (voting) model get on the ModelNet40 dataset
| 94.1±0.1 | Title: Point-JEPA: A Joint Embedding Predictive Architecture for Self-Supervised Learning on Point Cloud
Abstract: AbstractRecent advancements in self-supervised learning in the point cloud domain have demonstrated significant potential. However, these methods often suffer from drawbacks, including lengthy pre-trainin... | 2024-2025 | public |
CC3M-TagMask | TTD (w/o fine-tuning) | TTD: Text-Tag Self-Distillation Enhancing Image-Text Alignment in CLIP to Alleviate Single Tag Bias | 2024-03-30T00:00:00 | https://arxiv.org/abs/2404.00384v2 | [
"https://github.com/shjo-april/TTD"
] | In the paper 'TTD: Text-Tag Self-Distillation Enhancing Image-Text Alignment in CLIP to Alleviate Single Tag Bias', what F1 score did the TTD (w/o fine-tuning) model get on the CC3M-TagMask dataset
| 78.5 | Title: TTD: Text-Tag Self-Distillation Enhancing Image-Text Alignment in CLIP to Alleviate Single Tag Bias
Abstract: AbstractWe identify a critical bias in contemporary CLIP-based models, which we denote assingle tag bias. This bias manifests as a disproportionate focus on a singular tag (word) while neglecting other ... | 2024-2025 | public |
VATEX | CoCap (ViT/L14) | Accurate and Fast Compressed Video Captioning | 2023-09-22T00:00:00 | https://arxiv.org/abs/2309.12867v2 | [
"https://github.com/acherstyx/CoCap"
] | In the paper 'Accurate and Fast Compressed Video Captioning', what BLEU-4 score did the CoCap (ViT/L14) model get on the VATEX dataset
| 35.8 | Title: Accurate and Fast Compressed Video Captioning
Abstract: AbstractExisting video captioning approaches typically require to first sample video frames from a decoded video and then conduct a subsequent process (e.g., feature extraction and/or captioning model learning). In this pipeline, manual frame sampling may ... | 2023 | public |
GSM8K | OpenMath-CodeLlama-34B (w/ code, SC, k=50) | OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset | 2024-02-15T00:00:00 | https://arxiv.org/abs/2402.10176v2 | [
"https://github.com/kipok/nemo-skills"
] | In the paper 'OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset', what Accuracy score did the OpenMath-CodeLlama-34B (w/ code, SC, k=50) model get on the GSM8K dataset
| 88.0 | Title: OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset
Abstract: AbstractRecent work has shown the immense potential of synthetically generated datasets for training large language models (LLMs), especially for acquiring targeted skills. Current large-scale math instruction tuning datasets such as Me... | 2024-2025 | public |
Yelp2018 | NESCL | Neighborhood-Enhanced Supervised Contrastive Learning for Collaborative Filtering | 2024-02-18T00:00:00 | https://arxiv.org/abs/2402.11523v1 | [
"https://github.com/PeiJieSun/NESCL"
] | In the paper 'Neighborhood-Enhanced Supervised Contrastive Learning for Collaborative Filtering', what NDCG@20 score did the NESCL model get on the Yelp2018 dataset
| 0.0611 | Title: Neighborhood-Enhanced Supervised Contrastive Learning for Collaborative Filtering
Abstract: AbstractWhile effective in recommendation tasks, collaborative filtering (CF) techniques face the challenge of data sparsity. Researchers have begun leveraging contrastive learning to introduce additional self-supervised... | 2024-2025 | public |
WikiText-103 | Skip Cross-Head Transformer-XL | Memory-efficient Stochastic methods for Memory-based Transformers | 2023-11-14T00:00:00 | https://arxiv.org/abs/2311.08123v1 | [
"https://github.com/vishwajit-vishnu/memory-efficient-stochastic-methods-for-memory-based-transformers"
] | In the paper 'Memory-efficient Stochastic methods for Memory-based Transformers', what Validation perplexity score did the Skip Cross-Head Transformer-XL model get on the WikiText-103 dataset
| 21.87 | Title: Memory-efficient Stochastic methods for Memory-based Transformers
Abstract: AbstractTraining Memory-based transformers can require a large amount of memory and can be quite inefficient. We propose a novel two-phase training mechanism and a novel regularization technique to improve the training efficiency of mem... | 2023 | public |
Bongard-OpenWorld | InstructBLIP + GPT-4 | Bongard-OpenWorld: Few-Shot Reasoning for Free-form Visual Concepts in the Real World | 2023-10-16T00:00:00 | https://arxiv.org/abs/2310.10207v5 | [
"https://github.com/joyjayng/Bongard-OpenWorld"
] | In the paper 'Bongard-OpenWorld: Few-Shot Reasoning for Free-form Visual Concepts in the Real World', what 2-Class Accuracy score did the InstructBLIP + GPT-4 model get on the Bongard-OpenWorld dataset
| 63.8 | Title: Bongard-OpenWorld:Few-Shot Reasoning for Free-form Visual Concepts in the Real World
Abstract: AbstractWe introduceBongard-OpenWorld, a new benchmark for evaluating real-world few-shot reasoning for machine vision. It originates from the classicalBongard Problems (BPs): Given two sets of images (positive and ne... | 2023 | public |
WHAM! | SepReformer-L + DM | Separate and Reconstruct: Asymmetric Encoder-Decoder for Speech Separation | 2024-06-10T00:00:00 | https://arxiv.org/abs/2406.05983v3 | [
"https://github.com/dmlguq456/SepReformer"
] | In the paper 'Separate and Reconstruct: Asymmetric Encoder-Decoder for Speech Separation', what SI-SDRi score did the SepReformer-L + DM model get on the WHAM! dataset
| 18.4 | Title: Separate and Reconstruct:Asymmetric Encoder-Decoder for Speech Separation
Abstract: AbstractSince the recent success of a time-domain speech separation, further improvements have been made by expanding the length and channel of a feature sequence to increase the amount of computation. When temporally expanded t... | 2024-2025 | public |
FSS-1000 (1-shot) | GF-SAM | Bridge the Points: Graph-based Few-shot Segment Anything Semantically | 2024-10-09T00:00:00 | https://arxiv.org/abs/2410.06964v2 | [
"https://github.com/ANDYZAQ/GF-SAM"
] | In the paper 'Bridge the Points: Graph-based Few-shot Segment Anything Semantically', what Mean IoU score did the GF-SAM model get on the FSS-1000 (1-shot) dataset
| 88 | Title: Bridge the Points: Graph-based Few-shotSegment Anything Semantically
Abstract: AbstractThe recent advancements in large-scale pre-training techniques have significantly enhanced the capabilities of vision foundation models, notably the Segment Anything Model (SAM), which can generate precise masks based on poin... | 2024-2025 | public |
VoxCeleb2 | RTFS-Net-4 | RTFS-Net: Recurrent Time-Frequency Modelling for Efficient Audio-Visual Speech Separation | 2023-09-29T00:00:00 | https://arxiv.org/abs/2309.17189v4 | [
"https://github.com/spkgyk/RTFS-Net"
] | In the paper 'RTFS-Net: Recurrent Time-Frequency Modelling for Efficient Audio-Visual Speech Separation', what SI-SNRi score did the RTFS-Net-4 model get on the VoxCeleb2 dataset
| 11.5 | Title: RTFS-Net: Recurrent time-frequency modelling for efficient audio-visual speech separation
Abstract: AbstractAudio-visual speech separation methods aim to integrate different modalities to generate high-quality separated speech, thereby enhancing the performance of downstream tasks such as speech recognition. Mo... | 2023 | public |
Weather (720) | MoLE-DLinear | Mixture-of-Linear-Experts for Long-term Time Series Forecasting | 2023-12-11T00:00:00 | https://arxiv.org/abs/2312.06786v3 | [
"https://github.com/rogerni/mole"
] | In the paper 'Mixture-of-Linear-Experts for Long-term Time Series Forecasting', what MSE score did the MoLE-DLinear model get on the Weather (720) dataset
| 0.314 | Title: Mixture-of-Linear-Experts for Long-term Time Series Forecasting
Abstract: AbstractLong-term time series forecasting (LTSF) aims to predict future values of a time series given the past values. The current state-of-the-art (SOTA) on this problem is attained in some cases bylinear-centricmodels, which primarily f... | 2023 | public |
VGGSound | EquiAV | EquiAV: Leveraging Equivariance for Audio-Visual Contrastive Learning | 2024-03-14T00:00:00 | https://arxiv.org/abs/2403.09502v2 | [
"https://github.com/jongsuk1/equiav"
] | In the paper 'EquiAV: Leveraging Equivariance for Audio-Visual Contrastive Learning', what Top 1 Accuracy score did the EquiAV model get on the VGGSound dataset
| 67.1 | Title: EquiAV: Leveraging Equivariance for Audio-Visual Contrastive Learning
Abstract: AbstractRecent advancements in self-supervised audio-visual representation learning have demonstrated its potential to capture rich and comprehensive representations. However, despite the advantages of data augmentation verified in ... | 2024-2025 | public |
YouTube-VOS 2018 | DMT | Deficiency-Aware Masked Transformer for Video Inpainting | 2023-07-17T00:00:00 | https://arxiv.org/abs/2307.08629v1 | [
"https://github.com/yeates/dmt"
] | In the paper 'Deficiency-Aware Masked Transformer for Video Inpainting', what PSNR score did the DMT model get on the YouTube-VOS 2018 dataset
| 34.27 | Title: Deficiency-Aware Masked Transformer for Video Inpainting
Abstract: AbstractRecent video inpainting methods have made remarkable progress by utilizing explicit guidance, such as optical flow, to propagate cross-frame pixels. However, there are cases where cross-frame recurrence of the masked video is not availab... | 2023 | public |
CHILI-3K | GAT | CHILI: Chemically-Informed Large-scale Inorganic Nanomaterials Dataset for Advancing Graph Machine Learning | 2024-02-20T00:00:00 | https://arxiv.org/abs/2402.13221v2 | [
"https://github.com/UlrikFriisJensen/CHILI"
] | In the paper 'CHILI: Chemically-Informed Large-scale Inorganic Nanomaterials Dataset for Advancing Graph Machine Learning', what F1-score (Weighted) score did the GAT model get on the CHILI-3K dataset
| 0.461 +/- 0.000 | Title: CHILI:Chemically-InformedLarge-scaleInorganic Nanomaterials Dataset for Advancing Graph Machine Learning
Abstract: Abstract.Advances in graph machine learning (ML) have been driven by applications in chemistry as graphs have remained the most expressive representations of molecules. This has led to progress wit... | 2024-2025 | public |
ARC (Challenge) | LLaMA 3 8B + MoSLoRA (fine-tuned) | Mixture-of-Subspaces in Low-Rank Adaptation | 2024-06-16T00:00:00 | https://arxiv.org/abs/2406.11909v3 | [
"https://github.com/wutaiqiang/moslora"
] | In the paper 'Mixture-of-Subspaces in Low-Rank Adaptation', what Accuracy score did the LLaMA 3 8B + MoSLoRA (fine-tuned) model get on the ARC (Challenge) dataset
| 81.5 | Title: Mixture-of-Subspaces in Low-Rank Adaptation
Abstract: AbstractIn this paper, we introduce asubspace-inspired Low-Rank Adaptation (LoRA) method, which is computationally efficient, easy to implement, and readily applicable to large language, multimodal, and diffusion models. Initially, we equivalently decompose ... | 2024-2025 | public |
LibriTTS | RFWave | RFWave: Multi-band Rectified Flow for Audio Waveform Reconstruction | 2024-03-08T00:00:00 | https://arxiv.org/abs/2403.05010v3 | [
"https://github.com/bfs18/rfwave"
] | In the paper 'RFWave: Multi-band Rectified Flow for Audio Waveform Reconstruction', what PESQ score did the RFWave model get on the LibriTTS dataset
| 4.228 | Title: RFWave: Multi-band Rectified Flow for Audio Waveform Reconstruction
Abstract: AbstractRecent advancements in generative modeling have led to significant progress in audio waveform reconstruction from diverse representations. Although diffusion models have been used for reconstructing audio waveforms, they tend ... | 2024-2025 | public |
MAWPS | OpenMath-CodeLlama-70B (w/ code) | OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset | 2024-02-15T00:00:00 | https://arxiv.org/abs/2402.10176v2 | [
"https://github.com/kipok/nemo-skills"
] | In the paper 'OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset', what Accuracy (%) score did the OpenMath-CodeLlama-70B (w/ code) model get on the MAWPS dataset
| 95.7 | Title: OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset
Abstract: AbstractRecent work has shown the immense potential of synthetically generated datasets for training large language models (LLMs), especially for acquiring targeted skills. Current large-scale math instruction tuning datasets such as Me... | 2024-2025 | public |
CIFAR-10-LT (ρ=100) | FBL (ResNet-32) | Feature-Balanced Loss for Long-Tailed Visual Recognition | 2023-05-18T00:00:00 | https://arxiv.org/abs/2305.10772v1 | [
"https://github.com/juyongjiang/fbl"
] | In the paper 'Feature-Balanced Loss for Long-Tailed Visual Recognition', what Error Rate score did the FBL (ResNet-32) model get on the CIFAR-10-LT (ρ=100) dataset
| 17.54 | Title: Feature-Balanced Loss for Long-Tailed Visual Recognition
Abstract: AbstractDeep neural networks frequently suffer from performance degradation when the training data is long-tailed because several majority classes dominate the training, resulting in a biased model. Recent studies have made a great effort in sol... | 2023 | public |
VisA | FAIRnoDTD | FAIR: Frequency-aware Image Restoration for Industrial Visual Anomaly Detection | 2023-09-13T00:00:00 | https://arxiv.org/abs/2309.07068v1 | [
"https://github.com/liutongkun/fair"
] | In the paper 'FAIR: Frequency-aware Image Restoration for Industrial Visual Anomaly Detection', what Detection AUROC score did the FAIRnoDTD model get on the VisA dataset
| 97.1 | Title: FAIR: Frequency-aware Image Restoration for Industrial Visual Anomaly Detection
Abstract: AbstractImage reconstruction-based anomaly detection models are widely explored in industrial visual inspection. However, existing models usually suffer from the trade-off between normal reconstruction fidelity and abnorma... | 2023 | public |
J-HMDB | SgMg (Video-Swin-B) | Spectrum-guided Multi-granularity Referring Video Object Segmentation | 2023-07-25T00:00:00 | https://arxiv.org/abs/2307.13537v1 | [
"https://github.com/bo-miao/sgmg"
] | In the paper 'Spectrum-guided Multi-granularity Referring Video Object Segmentation', what Precision@0.5 score did the SgMg (Video-Swin-B) model get on the J-HMDB dataset
| 0.972 | Title: Spectrum-guided Multi-granularity Referring Video Object Segmentation
Abstract: AbstractCurrent referring video object segmentation (R-VOS) techniques extract conditional kernels from encoded (low-resolution) vision-language features to segment the decoded high-resolution features. We discovered that this cause... | 2023 | public |
Cora with Public Split: fixed 20 nodes per class | GAT+PGN | The Split Matters: Flat Minima Methods for Improving the Performance of GNNs | 2023-06-15T00:00:00 | https://arxiv.org/abs/2306.09121v1 | [
"https://github.com/foisunt/fmms-in-gnns"
] | In the paper 'The Split Matters: Flat Minima Methods for Improving the Performance of GNNs', what Accuracy score did the GAT+PGN model get on the Cora with Public Split: fixed 20 nodes per class dataset
| 83.26 ± 0.69% | Title: The Split Matters: Flat Minima Methods for Improving the Performance of GNNs
Abstract: AbstractWhen training a Neural Network, it is optimized using the available training data with the hope that it generalizes well to new or unseen testing data. At the same absolute value, a flat minimum in the loss landscape ... | 2023 | public |
VisA | TransFusion | TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection | 2023-11-16T00:00:00 | https://arxiv.org/abs/2311.09999v2 | [
"https://github.com/maticfuc/eccv_transfusion"
] | In the paper 'TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection', what Detection AUROC score did the TransFusion model get on the VisA dataset
| 98.7 | Title: TransFusion – A Transparency-Based Diffusion Model for Anomaly Detection
Abstract: AbstractSurface anomaly detection is a vital component in manufacturing inspection. Reconstructive anomaly detection methods restore the normal appearance of an object, ideally modifying only the anomalous regions. Due to the lim... | 2023 | public |
OoDIS | Mask2Anomaly | Unmasking Anomalies in Road-Scene Segmentation | 2023-07-25T00:00:00 | https://arxiv.org/abs/2307.13316v1 | [
"https://github.com/shyam671/mask2anomaly-unmasking-anomalies-in-road-scene-segmentation"
] | In the paper 'Unmasking Anomalies in Road-Scene Segmentation', what AP score did the Mask2Anomaly model get on the OoDIS dataset
| 1.24 | Title: Unmasking Anomalies in Road-Scene Segmentation
Abstract: AbstractAnomaly segmentation is a critical task for driving applications, and it is approached traditionally as a per-pixel classification problem. However, reasoning individually about each pixel without considering their contextual semantics results in ... | 2023 | public |
MVTec AD | MSFlow | MSFlow: Multi-Scale Flow-based Framework for Unsupervised Anomaly Detection | 2023-08-29T00:00:00 | https://arxiv.org/abs/2308.15300v1 | [
"https://github.com/cool-xuan/msflow"
] | In the paper 'MSFlow: Multi-Scale Flow-based Framework for Unsupervised Anomaly Detection', what Detection AUROC score did the MSFlow model get on the MVTec AD dataset
| 99.7 | Title: MSFlow: Multi-Scale Flow-based Framework for Unsupervised Anomaly Detection
Abstract: AbstractUnsupervised anomaly detection (UAD) attracts a lot of research interest and drives widespread applications, where only anomaly-free samples are available for training. Some UAD applications intend to locate the anomal... | 2023 | public |
MTL-AQA | FineParser | FineParser: A Fine-grained Spatio-temporal Action Parser for Human-centric Action Quality Assessment | 2024-05-11T00:00:00 | https://arxiv.org/abs/2405.06887v1 | [
"https://github.com/pku-icst-mipl/fineparser_cvpr2024"
] | In the paper 'FineParser: A Fine-grained Spatio-temporal Action Parser for Human-centric Action Quality Assessment', what Spearman Correlation score did the FineParser model get on the MTL-AQA dataset
| 95.85 | Title: FineParser: A Fine-grained Spatio-temporal Action Parser for Human-centric Action Quality Assessment
Abstract: AbstractExisting action quality assessment (AQA) methods mainly learn deep representations at the video level for scoring diverse actions. Due to the lack of a fine-grained understanding of actions in ... | 2024-2025 | public |
SMAC 3s5z_vs_4s6z | DMIX | A Unified Framework for Factorizing Distributional Value Functions for Multi-Agent Reinforcement Learning | 2023-06-04T00:00:00 | https://arxiv.org/abs/2306.02430v1 | [
"https://github.com/j3soon/dfac-extended"
] | In the paper 'A Unified Framework for Factorizing Distributional Value Functions for Multi-Agent Reinforcement Learning', what Median Win Rate score did the DMIX model get on the SMAC 3s5z_vs_4s6z dataset
| 83.52 | Title: A Unified Framework for Factorizing Distributional Value Functions for Multi-Agent Reinforcement Learning
Abstract: AbstractIn fully cooperative multi-agent reinforcement learning (MARL) settings, environments are highly stochastic due to the partial observability of each agent and the continuously changing pol... | 2023 | public |
ICDAR2015 | CLIP4STR-B | CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model | 2023-05-23T00:00:00 | https://arxiv.org/abs/2305.14014v3 | [
"https://github.com/VamosC/CLIP4STR"
] | In the paper 'CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model', what Accuracy score did the CLIP4STR-B model get on the ICDAR2015 dataset
| 90.6 | Title: CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model
Abstract: AbstractPre-trained vision-language models (VLMs) are the de-facto foundation models for various downstream tasks. However, scene text recognition methods still prefer backbones pre-trained on a single modali... | 2023 | public |
RST-DT | Bottom-up Llama 2 (13B) | Can we obtain significant success in RST discourse parsing by using Large Language Models? | 2024-03-08T00:00:00 | https://arxiv.org/abs/2403.05065v1 | [
"https://github.com/nttcslab-nlp/rstparser_eacl24"
] | In the paper 'Can we obtain significant success in RST discourse parsing by using Large Language Models?', what Standard Parseval (Span) score did the Bottom-up Llama 2 (13B) model get on the RST-DT dataset
| 78.3 | Title: Can we obtain significant success in RST discourse parsingby using Large Language Models?
Abstract: AbstractRecently, decoder-only pre-trained large language models (LLMs), with several tens of billion parameters, have significantly impacted a wide range of natural language processing (NLP) tasks. While encoder... | 2024-2025 | public |
MassSpecGym | Precursor m/z | MassSpecGym: A benchmark for the discovery and identification of molecules | 2024-10-30T00:00:00 | https://arxiv.org/abs/2410.23326v1 | [
"https://github.com/pluskal-lab/massspecgym"
] | In the paper 'MassSpecGym: A benchmark for the discovery and identification of molecules', what Hit Rate @ 1 score did the Precursor m/z model get on the MassSpecGym dataset
| 2.09 | Title: MassSpecGym: A benchmark for the discovery and identification of molecules
Abstract: AbstractThe discovery and identification of molecules in biological and environmental samples is crucial for advancing biomedical and chemical sciences. Tandem mass spectrometry (MS/MS) is the leading technique for high-through... | 2024-2025 | public |
WebQuestions | PaLM 2-M (one-shot) | PaLM 2 Technical Report | 2023-05-17T00:00:00 | https://arxiv.org/abs/2305.10403v3 | [
"https://github.com/eternityyw/tram-benchmark"
] | In the paper 'PaLM 2 Technical Report', what EM score did the PaLM 2-M (one-shot) model get on the WebQuestions dataset
| 26.9 | Title: PaLM 2 Technical Report
Abstract: AbstractWe introduce PaLM 2, a new state-of-the-art language model that has better multilingual
and reasoning capabilities and is more compute-efficient than its
predecessor PaLM. PaLM 2 is a Transformer-based model trained using a mixture of objectives. Through extensive evalu... | 2023 | public |
VLCS | GMDG (ResNet-50) | Rethinking Multi-domain Generalization with A General Learning Objective | 2024-02-29T00:00:00 | https://arxiv.org/abs/2402.18853v1 | [
"https://github.com/zhaorui-tan/GMDG_cvpr2024"
] | In the paper 'Rethinking Multi-domain Generalization with A General Learning Objective', what Average Accuracy score did the GMDG (ResNet-50) model get on the VLCS dataset
| 79.2 | Title: Rethinking Multi-domain Generalization with A General Learning Objective
Abstract: AbstractMulti-domain generalization (mDG) is universally aimed to minimize the discrepancy between training and testing distributions to enhance marginal-to-label distribution mapping. However, existing mDG literature lacks a gen... | 2024-2025 | public |
ImageNet | Poly-SA-ViT-S | Which Transformer to Favor: A Comparative Analysis of Efficiency in Vision Transformers | 2023-08-18T00:00:00 | https://arxiv.org/abs/2308.09372v3 | [
"https://github.com/tobna/whattransformertofavor"
] | In the paper 'Which Transformer to Favor: A Comparative Analysis of Efficiency in Vision Transformers', what Top 1 Accuracy score did the Poly-SA-ViT-S model get on the ImageNet dataset
| 78.34% | Title: Which Transformer to Favor:A Comparative Analysis of Efficiency in Vision Transformers
Abstract: AbstractThe growing popularity of Vision Transformers as the go-to models for image classification has led to an explosion of architectural modifications claiming to be more efficient than the original ViT. However,... | 2023 | public |
FRMT (Chinese - Taiwan) | Google Translate | PaLM 2 Technical Report | 2023-05-17T00:00:00 | https://arxiv.org/abs/2305.10403v3 | [
"https://github.com/eternityyw/tram-benchmark"
] | In the paper 'PaLM 2 Technical Report', what BLEURT score did the Google Translate model get on the FRMT (Chinese - Taiwan) dataset
| 68.5 | Title: PaLM 2 Technical Report
Abstract: AbstractWe introduce PaLM 2, a new state-of-the-art language model that has better multilingual
and reasoning capabilities and is more compute-efficient than its
predecessor PaLM. PaLM 2 is a Transformer-based model trained using a mixture of objectives. Through extensive evalu... | 2023 | public |
CelebA-HQ 256x256 | BOSS | Bellman Optimal Stepsize Straightening of Flow-Matching Models | 2023-12-27T00:00:00 | https://arxiv.org/abs/2312.16414v3 | [
"https://github.com/nguyenngocbaocmt02/boss"
] | In the paper 'Bellman Optimal Stepsize Straightening of Flow-Matching Models', what clean-FID score did the BOSS model get on the CelebA-HQ 256x256 dataset
| 20.13 | Title: Bellman Optimal Stepsize Straightening of Flow-Matching Models
Abstract: AbstractFlow matching is a powerful framework for generating high-quality samples in various applications, especially image synthesis. However, the intensive computational demands of these models, especially during the finetuning process a... | 2023 | public |
MATH | Shepherd+Mistral-7B (SFT on MetaMATH + PRM RL+ PRM rerank, k=256) | Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations | 2023-12-14T00:00:00 | https://arxiv.org/abs/2312.08935v3 | [
"https://huggingface.co/datasets/peiyi9979/Math-Shepherd"
] | In the paper 'Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations', what Accuracy score did the Shepherd+Mistral-7B (SFT on MetaMATH + PRM RL+ PRM rerank, k=256) model get on the MATH dataset
| 43.5 | Title: \methodname: Verify and Reinforce LLMs Step-by-step without Human Annotations
Abstract: AbstractIn this paper, we present an innovative process-oriented math process reward model called\methodname, which assigns a reward score to each step of math problem solutions. The training of\methodnameis achieved using a... | 2023 | public |
BIG-bench (Formal Fallacies Syllogisms Negation) | PaLM 2 (few-shot, k=3, CoT) | PaLM 2 Technical Report | 2023-05-17T00:00:00 | https://arxiv.org/abs/2305.10403v3 | [
"https://github.com/eternityyw/tram-benchmark"
] | In the paper 'PaLM 2 Technical Report', what Accuracy score did the PaLM 2 (few-shot, k=3, CoT) model get on the BIG-bench (Formal Fallacies Syllogisms Negation) dataset
| 57.2 | Title: PaLM 2 Technical Report
Abstract: AbstractWe introduce PaLM 2, a new state-of-the-art language model that has better multilingual
and reasoning capabilities and is more compute-efficient than its
predecessor PaLM. PaLM 2 is a Transformer-based model trained using a mixture of objectives. Through extensive evalu... | 2023 | public |
Baidu Mall | AnyLoc-VLAD-DINOv2 | AnyLoc: Towards Universal Visual Place Recognition | 2023-08-01T00:00:00 | https://arxiv.org/abs/2308.00688v2 | [
"https://github.com/AnyLoc/AnyLoc"
] | In the paper 'AnyLoc: Towards Universal Visual Place Recognition', what Recall@1 score did the AnyLoc-VLAD-DINOv2 model get on the Baidu Mall dataset
| 75.22 | Title: AnyLoc: Towards Universal Visual Place Recognitionhttps://anyloc.github.io/
Abstract: AbstractVisual Place Recognition (VPR) is vital for robot localization. To date, the most performant VPR approaches areenvironment- and task-specific: while they exhibit strong performance in structured environments (predomina... | 2023 | public |
PECC | Mixtral-8x7B-Instruct | PECC: Problem Extraction and Coding Challenges | 2024-04-29T00:00:00 | https://arxiv.org/abs/2404.18766v1 | [
"https://github.com/hallerpatrick/pecc"
] | In the paper 'PECC: Problem Extraction and Coding Challenges', what Pass@3 score did the Mixtral-8x7B-Instruct model get on the PECC dataset
| 8.35 | Title: PECC: Problem Extraction and Coding Challenges
Abstract: AbstractRecent advancements in large language models (LLMs) have showcased their exceptional abilities across various tasks, such as code generation, problem-solving and reasoning. Existing benchmarks evaluate tasks in isolation, yet the extent to which L... | 2024-2025 | public |
Fashion IQ | CaLa | CaLa: Complementary Association Learning for Augmenting Composed Image Retrieval | 2024-05-29T00:00:00 | https://arxiv.org/abs/2405.19149v2 | [
"https://github.com/chiangsonw/cala"
] | In the paper 'CaLa: Complementary Association Learning for Augmenting Composed Image Retrieval', what (Recall@10+Recall@50)/2 score did the CaLa model get on the Fashion IQ dataset
| 57.96 | Title: CaLa: Complementary Association Learning for Augmenting Composed Image Retrieval
Abstract: Abstract.Composed image retrieval (CIR) is the task of searching target images using an image-text pair as a query. Given the straightforward relation of query pair-target image, the dominant methods follow the learning p... | 2024-2025 | public |
ADE20K-150 | SED | SED: A Simple Encoder-Decoder for Open-Vocabulary Semantic Segmentation | 2023-11-27T00:00:00 | https://arxiv.org/abs/2311.15537v2 | [
"https://github.com/xb534/sed"
] | In the paper 'SED: A Simple Encoder-Decoder for Open-Vocabulary Semantic Segmentation', what mIoU score did the SED model get on the ADE20K-150 dataset
| 35.2 | Title: SED: A Simple Encoder-Decoder for Open-Vocabulary Semantic Segmentation
Abstract: AbstractOpen-vocabulary semantic segmentation strives to distinguish pixels into different semantic groups from an open set of categories. Most existing methods explore utilizing pre-trained vision-language models, in which the ke... | 2023 | public |
AudioCaps | SLAM-AAC | SLAM-AAC: Enhancing Audio Captioning with Paraphrasing Augmentation and CLAP-Refine through LLMs | 2024-10-12T00:00:00 | https://arxiv.org/abs/2410.09503v1 | [
"https://github.com/X-LANCE/SLAM-LLM"
] | In the paper 'SLAM-AAC: Enhancing Audio Captioning with Paraphrasing Augmentation and CLAP-Refine through LLMs', what CIDEr score did the SLAM-AAC model get on the AudioCaps dataset
| 0.841 | Title: SLAM-AAC: Enhancing Audio Captioning with Paraphrasing Augmentation and CLAP-Refine through LLMs††thanks:Co-first author∗. Corresponding author†.
Abstract: AbstractAutomated Audio Captioning (AAC) aims to generate natural textual descriptions for input audio signals. Recent progress in audio pre-trained models ... | 2024-2025 | public |
Urban100 - 2x upscaling | DRCT | DRCT: Saving Image Super-resolution away from Information Bottleneck | 2024-03-31T00:00:00 | https://arxiv.org/abs/2404.00722v5 | [
"https://github.com/ming053l/drct"
] | In the paper 'DRCT: Saving Image Super-resolution away from Information Bottleneck', what PSNR score did the DRCT model get on the Urban100 - 2x upscaling dataset
| 34.54 | Title: DRCT: Saving Image Super-Resolution away from Information Bottleneck
Abstract: AbstractIn recent years, Vision Transformer-based approaches for low-level vision tasks have achieved widespread success. Unlike CNN-based models, Transformers are more adept at capturing long-range dependencies, enabling the reconst... | 2024-2025 | public |
MM-Vet | DynMOE-LLaVA | Dynamic Mixture of Experts: An Auto-Tuning Approach for Efficient Transformer Models | 2024-05-23T00:00:00 | https://arxiv.org/abs/2405.14297v3 | [
"https://github.com/lins-lab/dynmoe"
] | In the paper 'Dynamic Mixture of Experts: An Auto-Tuning Approach for Efficient Transformer Models', what GPT-4 score score did the DynMOE-LLaVA model get on the MM-Vet dataset
| 33.6 | 1:Input data𝐱𝐱\mathbf{x}, initial gating network parameters𝐖gsubscript𝐖𝑔\mathbf{W}_{g},𝐆𝐆\mathbf{G}, andτ𝜏\tau, expertsE1,⋯,EKsubscript𝐸1⋯subscript𝐸1E_{1},\cdots,E_{K}, start record routing flagflags𝑓𝑙𝑎subscript𝑔𝑠flag_{s}, finish record routing flagflagf𝑓𝑙𝑎subscript𝑔𝑓flag_{f}.2:MoE layer outpu... | 2024-2025 | public |
MLO-Cn2 | Climatology | Effective Benchmarks for Optical Turbulence Modeling | 2024-01-07T00:00:00 | https://arxiv.org/abs/2401.03573v1 | [
"https://github.com/cdjellen/otbench"
] | In the paper 'Effective Benchmarks for Optical Turbulence Modeling', what RMSE score did the Climatology model get on the MLO-Cn2 dataset
| 0.658 | Title: Effective Benchmarks for Optical Turbulence Modeling
Abstract: AbstractOptical turbulence presents a significant challenge for communication, directed energy, and imaging systems, especially in the atmospheric boundary layer. Effective modeling of optical turbulence strength is critical for the development and ... | 2024-2025 | public |
Stanford Drone | PPT | Progressive Pretext Task Learning for Human Trajectory Prediction | 2024-07-16T00:00:00 | https://arxiv.org/abs/2407.11588v1 | [
"https://github.com/isee-laboratory/ppt"
] | In the paper 'Progressive Pretext Task Learning for Human Trajectory Prediction', what ADE-8/12 @K = 20 score did the PPT model get on the Stanford Drone dataset
| 7.03 | Title: Progressive Pretext Task Learning forHuman Trajectory Prediction
Abstract: AbstractHuman trajectory prediction is a practical task of predicting the future positions of pedestrians on the road, which typically covers all temporal ranges from short-term to long-term within a trajectory. However, existing works a... | 2024-2025 | public |
CHASE_DB1 | MERIT-GCASCADE | G-CASCADE: Efficient Cascaded Graph Convolutional Decoding for 2D Medical Image Segmentation | 2023-10-24T00:00:00 | https://arxiv.org/abs/2310.16175v1 | [
"https://github.com/SLDGroup/G-CASCADE"
] | In the paper 'G-CASCADE: Efficient Cascaded Graph Convolutional Decoding for 2D Medical Image Segmentation', what F1 score score did the MERIT-GCASCADE model get on the CHASE_DB1 dataset
| 0.8267 | Title: G-CASCADE: Efficient Cascaded Graph Convolutional Decoding for 2D Medical Image Segmentation
Abstract: AbstractIn recent years, medical image segmentation has become an important application in the field of computer-aided diagnosis. In this paper, we are the first to propose a new graph convolution-based decode... | 2023 | public |
Charades-STA | UniMD+Sync. | UniMD: Towards Unifying Moment Retrieval and Temporal Action Detection | 2024-04-07T00:00:00 | https://arxiv.org/abs/2404.04933v2 | [
"https://github.com/yingsen1/unimd"
] | In the paper 'UniMD: Towards Unifying Moment Retrieval and Temporal Action Detection', what R@1 IoU=0.5 score did the UniMD+Sync. model get on the Charades-STA dataset
| 63.98 | Title: UniMD: Towards Unifying Moment Retrieval and Temporal Action Detection
Abstract: AbstractTemporal Action Detection (TAD) focuses on detecting pre-defined actions, while Moment Retrieval (MR) aims to identify the events described by open-ended natural language within untrimmed videos. Despite that they focus on ... | 2024-2025 | public |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.