ETR: Entropy Trend Reward for Efficient Chain-of-Thought Reasoning
arXiv:2604.05355v1 Announce Type: new Abstract: Chain-of-thought (CoT) reasoning improves large language model performance on complex tasks, but often produces excessively long and inefficient reasoning traces. …
Xuan Xiong, Huan Liu, Li Gu, Zhixiang Chi, Yue Qiu, Yuanhao Yu, Yang Wang
조회수 74회