Large Language Models have achieved strong performance on reasoning tasks with objective answers by generating step-by-step solutions, but diagnosing where a multi-step reasoning trace might fail remains difficult. Confidence estimation offers a diagnostic signal, yet existing methods are restricted to final answers or require internal model access. In this paper, we introduce Stepwise Confidence Attribution (SCA), a framework for closed-source LLMs that assigns step-level confidence based only on generated reasoning traces. SCA applies the Information Bottleneck principle: steps aligning with consensus structures across correct solutions receive high confidence, while deviations are flagged as potentially erroneous. We propose two complementary methods: (1) NIBS, a non-parametric IB approach measuring consistency without graph structures, and (2) GIBS, a graph-based IB model that learns subgraphs through a differentiable mask to capture logical variability. Extensive experiments on mathematical reasoning and multi-hop question answering show that SCA reliably identifies low-confidence steps strongly correlated with reasoning errors. Moreover, using step-level confidence to guide self-correction improves the correction success rate by up to 13.5\% over answer-level feedback.

翻译：大语言模型在生成逐步解答的客观答案推理任务中表现强劲，但诊断多步推理轨迹中可能失败的位置仍然困难。置信度估计提供了诊断信号，但现有方法局限于最终答案或需要模型内部访问。本文提出逐步置信度归因（SCA）框架，该框架专为闭源大语言模型设计，仅基于生成的推理轨迹即可分配步骤级置信度。SCA应用信息瓶颈原理：与正确解答中共识结构一致的步骤获得高置信度，而偏离步骤则被标记为潜在错误。我们提出两种互补方法：（1）NIBS，一种无需图结构的非参数化信息瓶颈方法，用于测量一致性；（2）GIBS，一种基于图的信息瓶颈模型，通过可微掩码学习子图以捕捉逻辑变异性。在数学推理和多跳问答上的大量实验表明，SCA能可靠识别与推理错误高度相关的低置信度步骤。此外，使用步骤级置信度引导自我修正，较之答案级反馈可将修正成功率提升高达13.5%。