James Cheng
2 indexed papers
Publications per year
Top categories
Frequent co-authors
Research Timeline
This paper introduces Numca and Hista, two novel techniques that significantly improve state value estimation for LLM reinforcement learning, addressing the instability of standard critic approaches.
This paper introduces Numca and Hista, two novel techniques that significantly improve state value estimation for LLM reinforcement learning, addressing the instability of standard critic approaches.
Papers
Hista and Numca: Estimate State Value Effectively for LLM Reinforcement Learning
Zizhe Chen, Jiqian Dong, Yizhou Tian, Garry Yang +3 more
This paper introduces Numca and Hista, two novel techniques that significantly improve state value estimation for LLM reinforcement learning, addressing the instability of standard critic approaches.