发表机构
Nokia Bell Labs（诺基亚贝尔实验室）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究内在自我修正的利弊权衡，通过29个模型上的实验发现修订可能提升也可能损害准确性，提出应将其视为策略而非无条件有益，并比较了三种运行时选择。
AI 中文摘要
内在自我修正要求语言模型在没有收到新的外部证据的情况下修改自己的答案。第二次遍历可以纠正错误，但也可能推翻原本正确的答案。我们通过在BoolQ、GSM8K和Corr2Cause上对29个开放权重LLM进行正确性转换追踪，研究了这一权衡。总体准确率可能掩盖显著不同的修订行为：例如，Llama-3.1-8B在GSM8K上提升了25.5个百分点，而细化将19.1%的初始正确答案变为错误。一项受控的BoolQ研究进一步表明，细化提示会改变恢复与损害之间的平衡。然后，我们比较了三种运行时选择：保留初始答案、始终接受修订、以及使用初始响应后可用信号选择性地调用修订。该比较识别了学习门控有用的设置以及其他更简单的无条件策略表现更好的设置。这些结果表明，应将内在自我修正视为一种修订策略，而非普遍有益的第二次遍历，并通过其恢复的修正和引入的错误来评估它。
英文摘要
Intrinsic self-correction asks a language model to revise its own answer without receiving new external evidence. A second pass can recover mistakes, but it can also overturn answers that were already correct. We study this trade-off across 29 open-weight LLMs on BoolQ, GSM8K, and Corr2Cause by tracking correctness transitions between initial and revised answers. Aggregate accuracy can conceal substantially different revision behavior: for example, Llama-3.1-8B improves by 25.5 percentage points on GSM8K, while refinement changes 19.1% of initially correct answers into wrong ones. A controlled BoolQ study further shows that refinement prompts shift the balance between recovery and harm. We then compare three runtime choices: keeping the initial answer, always accepting the revision, and selectively invoking revision using signals available after the initial response. The comparison identifies settings where learned gating is useful and others where a simpler unconditional policy performs better. These results suggest treating intrinsic self-correction as a revision policy rather than as a uniformly beneficial second pass, and evaluating it through both the corrections it recovers and the errors it introduces.
