Beyond GRPO and On-Policy Distillation: An Empirical Sparse-to-Dense Reward Principle for Language-Model Post-Training
Paper • 2605.12483 • Published • 8
XYX
xuyd16
AI & ML interests
None yet
Recent Activity
Organizations
None yet
upvoted a paper 3 days ago
upvoted a paper 13 days ago
upvoted 4
papers 30 days ago
OccuBench: Evaluating AI Agents on Real-World Professional Tasks via Language World Models
Paper • 2604.10866 • Published • 66
SpatialEvo: Self-Evolving Spatial Intelligence via Deterministic Geometric Environments
Paper • 2604.14144 • Published • 63
RationalRewards: Reasoning Rewards Scale Visual Generation Both Training and Test Time
Paper • 2604.11626 • Published • 101
Seedance 2.0: Advancing Video Generation for World Complexity
Paper • 2604.14148 • Published • 160
upvoted 3
papers 2 months ago
PACED: Distillation at the Frontier of Student Competence
Paper • 2603.11178 • Published • 4
Overconfident Errors Need Stronger Correction: Asymmetric Confidence Penalties for Reinforcement Learning
Paper • 2602.21420 • Published • 6
Flash-KMeans: Fast and Memory-Efficient Exact K-Means
Paper • 2603.09229 • Published • 82