通过语义保持干预量化大模型认知不确定性，提升可靠性。
ESI: Epistemic Uncertainty Quantification via Semantic-preserving Intervention for Large Language Models
- 基于因果视角，用语义不变干预检测输出变化来估计不确定性
- 在多个大模型和问答数据集上验证，效果与效率均优于现有方法
- 适合关注大模型可信度评估的研究者与应用开发者
不确定性量化（UQ）是提升模型可靠性的重要方向，但大语言模型（LLM）的不确定性量化仍具挑战。本文从因果角度建立LLM不确定性与其在语义保持干预下不变性之间的联系。基于此，提出一种新的灰箱不确定性量化方法，通过比较干预前后模型输出的变化来度量不确定性。理论分析表明，该方法能有效估计认知不确定性。大量实验覆盖多种大模型及多个问答数据集，结果表明该方法不仅有效性突出，且计算效率高。
原文摘要 · Abstract (English)
Uncertainty Quantification (UQ) is a promising approach to improve model reliability, yet quantifying the uncertainty of Large Language Models (LLMs) is non-trivial. In this work, we establish a connection between the uncertainty of LLMs and their invariance under semantic-preserving intervention from a causal perspective. Building on this foundation, we propose a novel grey-box uncertainty quantification method that measures the variation in model outputs before and after the semantic-preserving intervention. Through theoretical justification, we show that our method provides an effective estimate of epistemic uncertainty. Our extensive experiments, conducted across various LLMs and a variety of question-answering (QA) datasets, demonstrate that our method excels not only in terms of effectiveness but also in computational efficiency.
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。