Think or Not? Exploring Thinking Efficiency in Large Reasoning Models via an Information-Theoretic Lens

NeurIPSSpotlight2025

Authors: Xixian Yong, Xiao Zhou, Yingying Zhang, Jinlin Li, Yefeng Zheng, Xian Wu
Venue: NeurIPS 2025
Track: Spotlight

TL;DR

The recent rise of Large Reasoning Models (LRMs) has significantly improved multi-step reasoning performance, but often at the cost of generating excessively long reasoning chains…

Opening excerpt from the authors’ abstract. source

Read the paper

OpenReview

Topics

reasoning

← All NeurIPS 2025 Spotlight papers · Browse the whole archive