研究大模型置信度评分的可区分精度，揭示其实际阈值选择受限。
The Score Granularity Gap in Black-Box LLM Classification: A Comparative Study of Confidence Constructions
- 对比7种置信度构造方法，发现单次输出评分值域极窄
- 强模型多轮查询反而降低排序性能，弱模型则受益
- 提出部署时权衡计算成本与阈值精细度的实用建议
大型语言模型（LLMs）越来越多地作为黑箱分类器用于自动化决策流程，其中需依据置信度分数对不确定样本进行人工审核。该场景要求置信度分数具备可设置阈值的分辨率。现有研究关注置信度是否校准或排序有效，本文提出一个被忽视的部署关键问题：分数能以多细粒度进行阈值划分？我们通过在25个模型-数据集组合（9个LLM，3个基准）上对比7种置信度构建方式——从单一口语化数值到词元概率，再到多次查询聚合结果——发现：单次输出的置信度经正确转换为类别概率后，虽排序表现意外良好，但仅取少数离散值，导致可用阈值极为有限。我们揭示了不同构造方法如何扩大该差距，及其对应的推理开销与排序效果，特别是多轮查询对弱模型有帮助，却可能损害强模型的表现。最终给出可落地的部署指导。
原文摘要 · Abstract (English)
Large language models (LLMs) are increasingly deployed as black-box classifiers in pipelines that automate confident decisions and route uncertain ones to human review. Such selective prediction needs a confidence score that an operator can threshold at a chosen risk level. Prior work asks whether LLM confidence is well calibrated or well ranked; we ask a complementary, deployment-oriented question that has been largely overlooked: at what resolution can the score be thresholded? We call the answer the score granularity gap. Through a controlled comparison of seven ways to build a confidence score, from a single verbalized number, to token probabilities, to querying the model many times and combining the answers, across 25 model-dataset pairs (9 LLMs, 3 benchmarks), we find that single-shot verbalized confidence, once correctly converted to a class probability, ranks cases surprisingly well, yet takes only a handful of distinct values. It therefore offers an operator only a few coarse thresholds, no matter how well it ranks. We show which constructions widen this gap, at what inference cost, and with what effect on ranking, notably that multi-query aggregation helps weak models but can degrade already-strong ones. We translate these trade-offs into concrete deployment guidance.
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。