一作ACL 2026Findings· Findings of ACL
C o TJ udger: A Graph-Driven Framework for Automatic Evaluation of Chain-of-Thought Efficiency and Redundancy in LRM s
Siyi Li · Jiajun Shi · Shiwen Ni · Ge Zhang · Shuaimin Li · Shijian Wang · Zhoufutu Wen · Yizhi Li · Hamid Alinejad-Rokny · JIAHENG LIU · Min Yang · Wenhao Huang
摘要
Large Reasoning Models (LRMs) have demonstrated strong performance by producing extended Chain-of-Thought (CoT) traces before answering. However, this paradigm often induces over-reasoning: redundant calculations and circular self-verification that increase computational cost without improving ou...