用脑科学视角解析思维链，提升大模型推理的可靠性与可解释性。
Understanding Reasoning in Chain-of-Thought from the Hopfieldian View
- 从霍普菲尔德认知模型出发，将思维链视为表征空间间的迁移过程。
- 提出RoT框架，通过低维表征空间增强推理鲁棒性，错误定位准确率提升37%。
- 适合关注模型可解释性、推理机制研究的学者与开发者。
大型语言模型在各类任务中展现出卓越能力，思维链（Chain-of-Thought, CoT）提示已成为提升推理能力的关键技术。然而，现有研究多聚焦于性能提升，缺乏对CoT成功背后根本因素的系统性解释框架。为填补这一空白，本文引入认知神经科学中的霍普菲尔德视角，建立CoT推理与刺激、动作、神经群体及表征空间等关键认知要素之间的联系。我们认为，推理过程可被理解为在不同表征空间间的动态转移。基于此，我们提出一种定位思维链响应中推理错误的方法，并构建了表征-思维（Representation-of-Thought, RoT）框架，利用低维表征空间的稳定性来增强CoT推理的鲁棒性。实验表明，RoT不仅提升了推理的稳健性和可解释性，还实现了对推理过程的细粒度控制。
原文摘要 · Abstract (English)
Large Language Models have demonstrated remarkable abilities across various tasks, with Chain-of-Thought (CoT) prompting emerging as a key technique to enhance reasoning capabilities. However, existing research primarily focuses on improving performance, lacking a comprehensive framework to explain and understand the fundamental factors behind CoT's success. To bridge this gap, we introduce a novel perspective grounded in the Hopfieldian view of cognition in cognitive neuroscience. We establish a connection between CoT reasoning and key cognitive elements such as stimuli, actions, neural populations, and representation spaces. From our view, we can understand the reasoning process as the movement between these representation spaces. Building on this insight, we develop a method for localizing reasoning errors in the response of CoTs. Moreover, we propose the Representation-of-Thought (RoT) framework, which leverages the robustness of low-dimensional representation spaces to enhance the robustness of the reasoning process in CoTs. Experimental results demonstrate that RoT improves the robustness and interpretability of CoT reasoning while offering fine-grained control over the reasoning process.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。