推理让视觉语言模型更自信，但反而降低了判断可靠性。
The Cost of Reasoning: Chain-of-Thought Induces Overconfidence in Vision-Language Models
- 推理过程会诱导模型基于自身思路而非真实不确定性做判断
- 尽管准确率提升，但多数不确定度估计质量下降
- 用一致性判断代替概率估计更适合带推理的模型
视觉语言模型（VLMs）在高风险场景中应用日益广泛，可靠不确定性量化（UQ）与预测准确性同等重要。当前主流的链式思维（CoT）提示或推理训练模型虽普遍用于提升性能，但其对UQ的影响仍不明确。实验表明，推理虽能提高任务准确率，却往往降低多种不确定性估计的质量。我们发现主要原因在于隐式答案依赖：在生成最终答案前，推理轨迹已趋向结论，导致令牌概率反映的是与自身推理的一致性，而非对正确性的不确定性。这使模型产生过度自信。相比之下，基于一致性的评估方法在推理下依然稳健，甚至有所提升，是推理型VLM中实用的不确定性估计方案。
原文摘要 · Abstract (English)
Vision-language models (VLMs) are increasingly deployed in high-stakes settings where reliable uncertainty quantification (UQ) is as important as predictive accuracy. Extended reasoning via chain-of-thought (CoT) prompting or reasoning-trained models has become ubiquitous in modern VLM pipelines, yet its effect on UQ reliability remains poorly understood. Our results show that reasoning tends to degrade the quality of many uncertainty estimates, even when it improves task accuracy. We identify implicit answer conditioning as the primary mechanism: as reasoning traces converge on a conclusion before the final answer is generated, token probabilities increasingly reflect consistency with the model's own reasoning trace rather than uncertainty about correctness. In effect, the model becomes overconfident in its answer. In contrast, agreement-based consistency remains robust and often improves under reasoning, making it a practical choice for uncertainty estimation in reasoning-enabled VLMs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。