Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning Paper • 2506.01939 • Published Jun 2 • 176
Absolute Zero: Reinforced Self-play Reasoning with Zero Data Paper • 2505.03335 • Published May 6 • 180
FaceID-6M: A Large-Scale, Open-Source FaceID Customization Dataset Paper • 2503.07091 • Published Mar 10 • 3
COIG-P: A High-Quality and Large-Scale Chinese Preference Dataset for Alignment with Human Values Paper • 2504.05535 • Published Apr 7 • 44
COIG-P: A High-Quality and Large-Scale Chinese Preference Dataset for Alignment with Human Values Paper • 2504.05535 • Published Apr 7 • 44
COIG-P: A High-Quality and Large-Scale Chinese Preference Dataset for Alignment with Human Values Paper • 2504.05535 • Published Apr 7 • 44
SuperGPQA: Scaling LLM Evaluation across 285 Graduate Disciplines Paper • 2502.14739 • Published Feb 20 • 105