While LLMs have seen substantial improvement in reasoning capabilities, they also sometimes overthink, generating unnecessary reasoning steps, particularly under uncertainty, given ill-posed or ambiguous queries. We introduce statistically principled early stopping methods that monitor uncertainty signals during generation to mitigate this issue. Our first approach is parametric: it models inter-arrival times of uncertainty keywords as a renewal process and applies sequential testing for stopping. Our second approach is nonparametric and provides finite-sample guarantees on the probability of halting too early on well-posed queries. We conduct empirical evaluations on reasoning tasks across several domains and models. Our results indicate that uncertainty-aware early stopping can improve both efficiency and reliability in LLM reasoning, and we observe especially significant gains for math reasoning.
@article{arxiv.2602.13935,
title = {Statistical Early Stopping for Reasoning Models},
author = {Yangxinyu Xie and Tao Wang and Soham Mallick and Yan Sun and Georgy Noarov and Mengxin Yu and Tanwi Mallick and Weijie J. Su and Edgar Dobriban},
journal= {arXiv preprint arXiv:2602.13935},
year = {2026}
}