作者位次 #2/4ACL 2026Findings· Findings of ACL
Beyond High-Entropy Exploration: Correctness-Aware Low-Entropy Segment-Based Advantage Shaping for Reasoning LLM s
Xinzhu Chen · Xuesheng Li · Zhongxiang Sun · Weijie Yu
摘要
Reinforcement Learning with Verifiable Rewards (RLVR) has become a central approach for improving the reasoning ability of large language models. Recent work studies RLVR through token entropy, arguing that high-entropy tokens drive exploration and should receive stronger updates. However, they o...