提出新方法评估大模型问答置信度,更准更实用。
A Semantic-Sampling Framework for Evaluating Calibration in Open-Ended Question Answering
- 从模型生成答案中按语义分组,用频率当置信度
- 在三个基准上优于现有方法,且能诊断题目难易
- 适合评估真实场景下大模型的可靠性
校准衡量模型预测置信度与其实际准确率的一致性,对医疗、法律等高风险领域部署大语言模型至关重要。现有评估方法在开放问答场景中表现不足:基于logit的指标需限制输出格式和内部概率;口头置信度自我报告且常过度自信;采样方法依赖任务特定提取规则,缺乏明确的有限样本目标。本文提出Sem-ECE(语义采样期望校准误差)框架,通过从模型采样答案并按语义分类,以结果频次作为置信度。研究两种估计器:Sem₁-ECE(同样本自一致性评分)和Sem₂-ECE(保留样本变体,分离答案选择与置信评估)。证明二者渐近无偏,且在简单问题上一致,在复杂问题上差异显著,后者校准误差更小,其差距可作为题目难度诊断。在五个主流商业LLM的三个开放问答基准上的实验验证理论预测,显示Sem-ECE优于口头置信度和现有采样方法,且在无法获取内部概率时仍可补全基于logit的评估。
原文摘要 · Abstract (English)
Calibration measures whether a model's predicted confidence aligns with its empirical accuracy, and is central to the reliable deployment of large language models (LLMs) in high-stakes domains such as medicine and law. While much recent work focuses on improving LLM calibration, the equally important question of how to evaluate it in realistic settings remains underdeveloped. Open-ended question answering (QA), the most common deployment setting for modern LLMs, is where existing evaluation methods fall short: logit-based metrics need restricted output formats and internal probabilities; verbalized confidence is self-reported and often overconfident; and sampling-based methods rely on task-specific extraction rules without a clear finite-sample target. We introduce Sem-ECE (Semantic-Sampling Expected Calibration Error), a calibration evaluation framework for open-ended QA that samples answers from the model, groups them into semantic classes, and uses the resulting frequencies as confidence. We study two estimators within this framework: Sem$_1$-ECE, the same-sample self-consistency score, and Sem$_2$-ECE, a held-out variant that separates answer selection from confidence evaluation. We prove both are asymptotically unbiased, and further show that they agree on easy questions but diverge on hard ones with Sem$_2$ achieving strictly smaller calibration error, so their gap also serves as a diagnostic for question difficulty. Experiments on three open-ended QA benchmarks across five leading commercial LLMs match our theoretical predictions and show that Sem-ECE outperforms verbalized confidence and existing sampling-based methods, while complementing logit-based evaluation when internal probabilities are unavailable.
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。