让大模型的口头概率更准确，提升判断可信度。
Calibrating Verbalized Probabilities for Large Language Models
- 用反向softmax近似原始输出，解决概率失真问题。
- 在三个数据集上验证了生成概率分布的可靠性。
- 适合需要可信置信度的AI决策场景。
校准口头概率为可靠评估和利用黑箱大语言模型输出提供了新方法。近期研究通过应用Platt缩放或温度缩放等技术，提升了大模型生成置信度分数的校准性。本文探讨了判别任务中口头概率分布的校准问题。首先，我们研究了大模型生成类别标签概率分布的能力。理论上并实证发现，口头概率缩放会导致重新软最大化的偏差，为此提出使用反向softmax技巧，通过反转口头概率来近似原始logits。在三个公开数据集上的广泛评估表明：（1）大模型具备生成类别分布的稳健能力；（2）反向softmax技巧能有效估计logits，从而支持后续校准调整。
原文摘要 · Abstract (English)
Calibrating verbalized probabilities presents a novel approach for reliably assessing and leveraging outputs from black-box Large Language Models (LLMs). Recent methods have demonstrated improved calibration by applying techniques like Platt scaling or temperature scaling to the confidence scores generated by LLMs. In this paper, we explore the calibration of verbalized probability distributions for discriminative tasks. First, we investigate the capability of LLMs to generate probability distributions over categorical labels. We theoretically and empirically identify the issue of re-softmax arising from the scaling of verbalized probabilities, and propose using the invert softmax trick to approximate the "logit" by inverting verbalized probabilities. Through extensive evaluation on three public datasets, we demonstrate: (1) the robust capability of LLMs in generating class distributions, and (2) the effectiveness of the invert softmax trick in estimating logits, which, in turn, facilitates post-calibration adjustments.
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。