Large language models (LLMs) improve their performance in downstream tasks
when they generate Chain of Thought reasoning text before producing an answer.
Our research investigates how LLMs recover from errors in Chain of Thought,
reaching the correct final answer despite mistakes in the reasoning text.
Through analysis of these error recovery behaviors, we find evidence for
unfaithfulness in Chain of Thought, but we also identify many clear examples of
faithful error recovery behaviors. We identify factors that shift LLM recovery
behavior: LLMs recover more frequently from obvious errors and in contexts that
provide more evidence for the correct answer. However, unfaithful recoveries
show the opposite behavior, occurring more frequently for more difficult error
positions. Our results indicate that there are distinct mechanisms driving
faithful and unfaithful error recoveries. Our results challenge the view that
LLM reasoning is a uniform, coherent process.

大型语言模型（LLMs）在输出答案之前生成思维链推理文本可以提高其在下游任务中的性能。我们的研究调查了 LLMs 在思维链中出现错误时如何还原并最终得出正确答案，我们发现存在思维链不忠实的证据，但我们也找到了许多明确的忠实错误还原行为。我们确定了影响 LLM 还原行为的因素：LLMs 在明显的错误和能提供更多正确答案证据的情境中更频繁地还原。然而，不忠实的还原行为则表现相反，在更困难的错误位置更频繁地发生。我们的研究结果显示，驱动忠实和不忠实错误还原的机制是不同的，这对 LLM 的推理是一个统一、连贯过程的观点提出了挑战。