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.

翻译：尽管大型语言模型在推理能力方面取得了显著进步，但它们有时也会过度思考，生成不必要的推理步骤，特别是在面对定义不清或模糊查询时的不确定性情况下。我们提出了基于统计原理的早期停止方法，通过在生成过程中监测不确定性信号来缓解这一问题。我们的第一种方法是参数化的：它将不确定性关键词的到达间隔时间建模为更新过程，并应用序贯检验进行停止决策。第二种方法是非参数化的，为定义明确查询过早停止的概率提供了有限样本保证。我们在多个领域和模型的推理任务上进行了实证评估。结果表明，具有不确定性感知的早期停止能够同时提升大型语言模型推理的效率和可靠性，尤其在数学推理任务中我们观察到了尤为显著的性能提升。