arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2603.21286cs.HC
收藏

当链条断裂时:LLM推理错误的交互诊断

When the Chain Breaks: Interactive Diagnosis of LLM Chain-of-Thought Reasoning Errors

College of Computing and Data Science, Nanyang Technological University(南洋理工大学计算机与数据科学学院);Bosch Research North America(博世北美研究院)

Shiwei Chen, Niruthikka Sritharan, Xiaolin Wen, Chenxi Zhang, Xingbo Wang, Yong Wang

AI总结:

本文提出ReasonDiag系统,通过结合外部事实核查与符号逻辑验证,帮助用户高效识别和诊断LLM推理链条中的错误步骤及根本原因。

摘要:

当前大型语言模型(LLM),尤其是大型推理模型,能够生成推理轨迹(CoT)以说明其最终输出的生成过程,从而帮助用户校准信任。然而,这些CoT推理轨迹通常冗长繁琐,且可能包含逻辑和事实错误,使用户难以高效准确地解读。为此,我们开发了一个结合外部事实核查与符号形式逻辑验证的错误检测流水线,以在步骤层面识别错误。基于此流水线,我们提出了ReasonDiag,一个用于诊断CoT推理轨迹的交互可视化系统。ReasonDiag提供1)集成的弧图以展示推理步骤分布和错误传播模式,2)分层的节点链接图以可视化高级推理流程和前提依赖。我们通过技术评估、两个案例研究和16名参与者的用户访谈评估了ReasonDiag。结果表明,ReasonDiag帮助用户有效理解CoT推理轨迹,识别错误步骤并确定其根本原因。

英文摘要:

Current Large Language Models (LLMs), especially Large Reasoning Models, can generate Chain-of-Thought (CoT) reasoning traces to illustrate how they produce final outputs, thereby facilitating trust calibration for users. However, these CoT reasoning traces are usually lengthy and tedious, and can contain various issues, such as logical and factual errors, which make it difficult for users to interpret the reasoning traces efficiently and accurately. To address these challenges, we develop an error detection pipeline that combines external fact-checking with symbolic formal logical validation to identify errors at the step level. Building on this pipeline, we propose ReasonDiag, an interactive visualization system for diagnosing CoT reasoning traces. ReasonDiag provides 1) an integrated arc diagram to show reasoning-step distributions and error-propagation patterns, and 2) a hierarchical node-link diagram to visualize high-level reasoning flows and premise dependencies. We evaluate ReasonDiag through a technical evaluation for the error detection pipeline, two case studies, and user interviews with 16 participants. The results indicate that ReasonDiag helps users effectively understand CoT reasoning traces, identify erroneous steps, and determine their root causes.

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。

↑