让大模型先推理再判断，能更真实地反映它的信心程度。
Read Your Own Mind: Reasoning Helps Surface Self-Confidence Signals in LLMs
- 强制模型先生成长链推理再输出答案，提升信心评估准确性。
- 在简单问答中，推理后信心评分准确率显著提高。
- 适合研究大模型不确定性或构建可信AI系统的开发者。
我们分析了 DeepSeek R1-32B 在问答任务中自报信心的表现。默认的先答后评模式下，模型常过度自信；而通过采样大量回答计算出的语义熵则保持可靠。我们推测这是因为语义熵利用了更大的推理时计算量，可探索模型的预测分布。实验表明，强制模型在作答前进行长链推理，显著提升了其自报信心的有效性，即使在仅需事实检索的简单问题上也是如此。此外，仅观察推理链的独立阅读模型也能重建相似的信心评分，表明自报信心可能只是推理过程中浮现的选项统计结果。分析结论：可靠的不确定性估计需要显式探索生成空间，自报信心仅在充分探索后才可信。
原文摘要 · Abstract (English)
We study the source of uncertainty in DeepSeek R1-32B by analyzing its self-reported verbal confidence on question answering (QA) tasks. In the default answer-then-confidence setting, the model is regularly over-confident, whereas semantic entropy - obtained by sampling many responses - remains reliable. We hypothesize that this is because of semantic entropy's larger test-time compute, which lets us explore the model's predictive distribution. We show that granting DeepSeek the budget to explore its distribution by forcing a long chain-of-thought before the final answer greatly improves its verbal score effectiveness, even on simple fact-retrieval questions that normally require no reasoning. Furthermore, a separate reader model that sees only the chain can reconstruct very similar confidences, indicating the verbal score might be merely a statistic of the alternatives surfaced during reasoning. Our analysis concludes that reliable uncertainty estimation requires explicit exploration of the generative space, and self-reported confidence is trustworthy only after such exploration.
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。