SPICE: Self-Play In Corpus Environments Improves Reasoning Paper • 2510.24684 • Published Oct 28 • 15
Verbalized Sampling: How to Mitigate Mode Collapse and Unlock LLM Diversity Paper • 2510.01171 • Published Oct 1 • 18
BigCodeArena: Unveiling More Reliable Human Preferences in Code Generation via Execution Paper • 2510.08697 • Published Oct 9 • 35
EnvPool: A Highly Parallel Reinforcement Learning Environment Execution Engine Paper • 2206.10558 • Published Jun 21, 2022 • 1
Efficient Process Reward Model Training via Active Learning Paper • 2504.10559 • Published Apr 14 • 13