让大模型先估信心再作答，提升不确定性判断的准确性与实用性。
Confidence Before Answering: A Paradigm Shift for Efficient LLM Uncertainty Estimation
- 先输出信心分数，再生成答案，实现信心与回答协同优化。
- 在数学、编程和事实问答任务中，信心校准更准，区分能力更强。
- 适合需要可靠不确定性的场景，如医疗、金融等安全敏感领域。
大语言模型的可靠部署依赖于准确的不确定性估计。现有方法多为答案优先，仅在生成回答后才给出置信度，衡量的是特定回答的正确性，限制了实际应用。本文提出信心优先范式，模型在回答前输出信心分数，将其解释为当前策略下回答正确的概率。我们提出CoCA（Co-optimized Confidence and Answers）框架，基于分段信用分配的GRPO强化学习方法，联合优化信心校准与回答准确性。通过为信心和回答段落分别设置奖励和组内相对优势，CoCA实现了稳定联合优化，避免奖励劫持问题。在数学、代码和事实问答基准上的实验表明，该方法在保持回答质量的同时，显著提升了信心校准精度和不确定性判别能力，拓展了下游应用范围。
原文摘要 · Abstract (English)
Reliable deployment of large language models (LLMs) requires accurate uncertainty estimation. Existing methods are predominantly answer-first, producing confidence only after generating an answer, which measure the correctness of a specific response and limits practical usability. We study a confidence-first paradigm, where the model outputs its confidence before answering, interpreting this score as the model's probability of answering the question correctly under its current policy. We propose CoCA(Co-optimized Confidence and Answers), a GRPO reinforcement learning framework that jointly optimizes confidence calibration and answer accuracy via segmented credit assignment. By assigning separate rewards and group-relative advantages to confidence and answer segments, CoCA enables stable joint optimization and avoids reward hacking. Experiments across math, code, and factual QA benchmarks show improved calibration and uncertainty discrimination while preserving answer quality, thereby enabling a broader range of downstream applications.
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。