分析思维链动态：主动引导还是事后的不忠实合理化？

摘要

近期研究表明，在诸如分析推理和常识推理等软推理问题上，思维链（CoT）方法带来的提升往往有限。此外，CoT可能无法忠实反映模型的实际推理过程。我们探究了在软推理任务中，经过指令调优的模型、推理模型以及推理蒸馏模型使用CoT的动态特性及其忠实度。研究发现，这些模型对CoT的依赖方式存在差异，且CoT的影响与其忠实度并非总是一致。

English

Recent work has demonstrated that Chain-of-Thought (CoT) often yields limited gains for soft-reasoning problems such as analytical and commonsense reasoning. CoT can also be unfaithful to a model's actual reasoning. We investigate the dynamics and faithfulness of CoT in soft-reasoning tasks across instruction-tuned, reasoning and reasoning-distilled models. Our findings reveal differences in how these models rely on CoT, and show that CoT influence and faithfulness are not always aligned.

分析思维链动态：主动引导还是事后的不忠实合理化？

Analysing Chain of Thought Dynamics: Active Guidance or Unfaithful Post-hoc Rationalisation?

摘要

Support