通过自适应共形语义熵进行LLM不确定性量化
LLMs Uncertainty Quantification via Adaptive Conformal Semantic Entropy
AI总结：
提出自适应共形语义熵（ACSE）方法，通过聚类语义熵并自适应调整不确定性分数，结合共形校准实现统计可靠的接受/弃权决策，在多个数据集上优于现有基线。
摘要：
LLMs的过度自信，特别是在产生幻觉时，对在安全关键环境中部署模型构成了重大挑战，并使得对不确定性进行可靠估计成为必要。现有的不确定性量化方法通常优先考虑词汇或概率度量；然而，这些技术往往忽略了具有相似含义的不同响应的语义差异。在本文中，我们提出了自适应共形语义熵（ACSE），一种通过自适应测量LLMs输出中的语义分散性来估计提示级不确定性的方法。我们的不确定性评分函数基于对同一提示的多个不同响应的语义熵进行聚类。该函数根据每个聚类的语义特征自适应调整不确定性分数。为了确保我们分数的统计可靠性，我们使用共形校准应用决策规则来接受/弃权提示，提供了有限样本、无分布的保证，使得接受响应中的错误率保持在用户指定的容差范围内。我们使用不同LLMs和数据集进行的广泛实验评估表明，我们的方法在判别性能、共形保证和概率校准指标方面始终优于最先进的不确定性量化基线。作为一个亮点，对于TriviaQA数据集，我们方法的AUROC为0.88，而令牌熵方法为0.65。
英文摘要：
LLMs' overconfidence, particularly when hallucinating, poses a significant challenge for the deployment of the models in safety-critical settings and makes a reliable estimation of uncertainty necessary. Existing approaches for uncertainty quantification typically prioritize lexical or probabilistic measures; however, these techniques often ignore the semantic variance of different responses with similar meaning. In this paper, we propose Adaptive Conformal Semantic Entropy (ACSE), a method for estimating prompt-level uncertainty by adaptively measuring semantic dispersion in LLMs outputs. Our uncertainty scoring function is based on clustering semantic entropy of multiple diverse responses to the same prompt. The function adaptively adjusts the uncertainty score based on semantic features of each cluster. To ensure statistical reliability of our score, we use conformal calibration to apply a decision rule to accept/abstain the prompts, providing a finite-sample, distribution-free guarantee such that the error rate among the accepted responses remains bounded by a user-specified tolerance. Our extensive experimental evaluations using different LLMs and datasets, demonstrate that our approach consistently outperforms state-of-the-art uncertainty quantification baselines using discriminative performance, conformal guarantees, and probabilistic calibration indicators. As a highlight, for TriviaQA dataset, AUROC of our approach is 0.88 compared to 0.65 produced by the token entropy approach.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
