言语化置信度触发自我验证：无需显式推理监督的涌现行为
Verbalized Confidence Triggers Self-Verification: Emergent Behavior Without Explicit Reasoning Supervision
AI总结：
本文发现仅用标量置信度标签微调即可让大语言模型涌现自我验证行为，并提出基于校准不确定性的测试时扩展方法，在多个推理基准上同时提升校准性、准确性和可解释性。
摘要：
不确定性校准对于大语言模型（LLMs）的安全部署至关重要，尤其是当用户依赖言语化置信度估计时。此前的工作主要关注分类器或短文本生成，而针对思维链（CoT）推理的置信度校准在很大程度上仍未得到探索。令人惊讶的是，我们发现仅使用标量置信度标签进行监督微调就足以引发语言模型的自我验证行为，无需任何显式推理监督或基于强化学习的奖励。尽管模型仅被训练为生成言语化置信度分数，且未接触任何自我验证示例，但它学会了在低置信度查询中生成更长且带有自我检查的响应，同时在高置信度查询中给出更简洁的回答。我们进一步提出了一种简单的重新思考方法，该方法基于校准后的不确定性通过测试时扩展来提升性能。在GSM8K以及MATH-500和ARC-Challenge等留出推理任务上的实验表明，我们的置信度感知微调同时改善了校准性和准确性，并通过使模型的推理路径与其置信度保持一致来增强可解释性。
英文摘要：
Uncertainty calibration is essential for the safe deployment of large language models (LLMs), particularly when users rely on verbalized confidence estimates. While prior work has focused on classifiers or short-form generation, confidence calibration for chain-of-thought (CoT) reasoning remains largely unexplored. Surprisingly, we find that supervised fine-tuning with scalar confidence labels alone suffices to elicit self-verification behavior of language models, without any explicit reasoning supervision or reinforcement learning-based rewards. Despite being trained only to produce a verbalized confidence score without any self-verifying examples, the model learns to generate longer and self-checking responses for low-confidence queries while providing more concise answers for high-confidence ones. We further propose a simple rethinking method that boosts performance via test-time scaling based on calibrated uncertainty. Experiments on GSM8K and held-out reasoning tasks such as MATH-500 and ARC-Challenge show that our confidence-aware fine-tuning improves both calibration and accuracy, while also enhancing interpretability by aligning the model's reasoning path with its confidence.
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。
