通过其计算图验证推理链
Verifying Chain-of-Thought Reasoning via Its Computational Graph
AI总结：
通过分析计算图结构特征，提出基于电路的推理验证方法，揭示推理错误的结构指纹并有效纠正模型错误推理。
摘要：
当前的推理链（CoT）验证方法基于输出（黑盒）或激活（灰盒）来预测推理的正确性，但对计算失败的原因缺乏深入理解。我们引入了一种白盒方法：基于电路的推理验证（CRV）。我们假设正确的推理步骤的归因图，作为模型潜在推理电路的执行轨迹，具有与错误步骤不同的结构指纹。通过在这些图的结构特征上训练分类器，我们证明这些轨迹包含强大的推理错误信号。我们的白盒方法产生了其他方法无法获得的科学见解。（1）我们证明错误的结构特征具有高度预测性，确立了通过其计算图直接验证推理的可行性。（2）我们发现这些特征具有高度领域特定性，揭示了不同推理任务中的失败表现为不同的计算模式。（3）我们提供了证据表明这些特征不仅仅是相关性；通过使用我们的分析指导对单个转码器特征的针对性干预，我们成功纠正了模型的错误推理。我们的工作表明，通过审视模型的计算过程，我们可以从简单的错误检测发展到对大语言模型推理的更深入的因果理解。
英文摘要：
Current Chain-of-Thought (CoT) verification methods predict reasoning correctness based on outputs (black-box) or activations (gray-box), but offer limited insight into why a computation fails. We introduce a white-box method: Circuit-based Reasoning Verification (CRV). We hypothesize that attribution graphs of correct CoT steps, viewed as execution traces of the model's latent reasoning circuits, possess distinct structural fingerprints from those of incorrect steps. By training a classifier on structural features of these graphs, we show that these traces contain a powerful signal of reasoning errors. Our white-box approach yields novel scientific insights unattainable by other methods. (1) We demonstrate that structural signatures of error are highly predictive, establishing the viability of verifying reasoning directly via its computational graph. (2) We find these signatures to be highly domain-specific, revealing that failures in different reasoning tasks manifest as distinct computational patterns. (3) We provide evidence that these signatures are not merely correlational; by using our analysis to guide targeted interventions on individual transcoder features, we successfully correct the model's faulty reasoning. Our work shows that, by scrutinizing a model's computational process, we can move from simple error detection to a deeper, causal understanding of LLM reasoning.
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。
