4 papers
Lookahead Tree-Based Rollouts for Enhanced Trajectory-Level Exploration in Reinforcement Learning with Verifiable Rewards
Shangyu Xing, Siyuan Wang, Chenyuan Yang +2
Reinforcement Learning with Verifiable Rewards (RLVR), particularly with algorithms like Group Relative Policy Optimization (GRPO), has proven highly effective in enhancing the rea…
Segment-Level Attribution for Selective Learning of Long Reasoning Traces
Siyuan Wang, Yanchen Liu, Xiang Ren
Large Reasoning Models (LRMs) achieve strong reasoning performance by generating long chains of thought (CoTs), yet only a small fraction of these traces meaningfully contributes t…
Diagnosing Memorization in Chain-of-Thought Reasoning, One Token at a Time
Huihan Li, You Chen, Siyuan Wang +4
Large Language Models (LLMs) perform well on reasoning benchmarks but often fail when inputs alter slightly, raising concerns about the extent to which their success relies on memo…
Stepwise Informativeness Search for Efficient and Effective LLM Reasoning
Siyuan Wang, Enda Zhao, Zhongyu Wei +1
Advances in Large Language Models (LLMs) have significantly improved multi-step reasoning through generating free-text rationales. However, recent studies show that LLMs tend to lo…