FaithCoT-Bench: 对链式推理实例级忠实度的基准测试
FaithCoT-Bench: Benchmarking Instance-Level Faithfulness of Chain-of-Thought Reasoning
AI总结：
FaithCoT-Bench提出了一种统一的基准，用于评估链式推理在实例级的忠实度，通过专家标注的轨迹和系统评估揭示了现有方法的优缺点。
摘要：
大型语言模型（LLMs）越来越多地依赖链式推理（CoT）提示来提高问题解决能力和提供看似透明的解释。然而，越来越多的证据表明，CoT往往无法忠实代表底层的推理过程，引发了对其在高风险应用中可靠性的问题。尽管先前研究集中在机制层面分析，显示CoT可能不忠实，但它们留有空白，即如何决定特定轨迹是否忠实于模型的内部推理过程。为解决这一差距，我们引入了FaithCoT-Bench，一个统一的基准，用于实例级CoT不忠实度检测。我们的框架建立了一个严谨的任务公式，将不忠实度检测制定为判别决策问题，并提供了FINE-CoT（链式推理实例忠实度评估），一个由四个代表性LLMs在四个领域生成的超过1000条轨迹的专家标注集合，其中包括超过300条不忠实实例，具有细粒度的原因和步骤级证据。我们进一步对十一种代表性检测方法进行了系统评估，涵盖了反事实、基于logit和LLM-as-judge范式，得出了经验洞察，澄清了现有方法的优势和劣势，并揭示了在知识密集型领域和更高级模型中的检测挑战增加。据我们所知，FaithCoT-Bench建立了首个全面的实例级CoT忠实度基准，为未来研究更可解释和可信的LLM推理奠定了坚实基础。
英文摘要：
Large language models (LLMs) increasingly rely on Chain-of-Thought (CoT) prompting to improve problem-solving and provide seemingly transparent explanations. However, growing evidence shows that CoT often fail to faithfully represent the underlying reasoning process, raising concerns about their reliability in high-risk applications. Although prior studies have focused on mechanism-level analyses showing that CoTs can be unfaithful, they leave open the practical challenge of deciding whether a specific trajectory is faithful to the internal reasoning of the model. To address this gap, we introduce FaithCoT-Bench, a unified benchmark for instance-level CoT unfaithfulness detection. Our framework establishes a rigorous task formulation that formulates unfaithfulness detection as a discriminative decision problem, and provides FINE-CoT (Faithfulness instance evaluation for Chain-of-Thought), an expert-annotated collection of over 1,000 trajectories generated by four representative LLMs across four domains, including more than 300 unfaithful instances with fine-grained causes and step-level evidence. We further conduct a systematic evaluation of eleven representative detection methods spanning counterfactual, logit-based, and LLM-as-judge paradigms, deriving empirical insights that clarify the strengths and weaknesses of existing approaches and reveal the increased challenges of detection in knowledge-intensive domains and with more advanced models. To the best of our knowledge, FaithCoT-Bench establishes the first comprehensive benchmark for instance-level CoT faithfulness, setting a solid basis for future research toward more interpretable and trustworthy reasoning in LLMs.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
