错误信念图谱：通过注意力模式追踪内在与外在幻觉
The Map of Misbelief: Tracing Intrinsic and Extrinsic Hallucinations Through Attention Patterns
AI总结：
针对LLMs幻觉检测中采样策略成本高、忽视幻觉类型区分的问题，提出区分内外在幻觉的评估框架，结合基于注意力的不确定性量化算法及新聚合策略，提升检测性能与可解释性，为幻觉检测提供新方向。
摘要：
大语言模型（LLMs）正越来越多地应用于安全关键领域，但仍容易产生幻觉。虽然先前的研究提出了用于幻觉检测的置信度表示方法，但这些方法大多依赖计算成本高昂的采样策略，且往往忽视幻觉类型的区分。在本研究中，我们引入了一个有原则的评估框架，该框架区分外在和内在幻觉类别，并在一系列精心设计的基准上评估检测性能。此外，我们利用了一种最新的基于注意力的不确定性量化算法，并提出了新的注意力聚合策略，以同时提升可解释性和幻觉检测性能。我们的实验结果表明，像语义熵这样的基于采样的方法能有效检测外在幻觉，但通常在内在幻觉上表现不佳；相比之下，我们的通过聚合输入标记注意力的方法更适合检测内在幻觉。这些见解为使检测策略与幻觉性质相匹配提供了新方向，并强调注意力是量化模型不确定性的丰富信号。
英文摘要：
Large Language Models (LLMs) are increasingly deployed in safety-critical domains, yet remain susceptible to hallucinations. While prior works have proposed confidence representation methods for hallucination detection, most of these approaches rely on computationally expensive sampling strategies and often disregard the distinction between hallucination types. In this work, we introduce a principled evaluation framework that differentiates between extrinsic and intrinsic hallucination categories and evaluates detection performance across a suite of curated benchmarks. In addition, we leverage a recent attention-based uncertainty quantification algorithm and propose novel attention aggregation strategies that improve both interpretability and hallucination detection performance. Our experimental findings reveal that sampling-based methods like Semantic Entropy are effective for detecting extrinsic hallucinations but generally fail on intrinsic ones. In contrast, our method, which aggregates attention over input tokens, is better suited for intrinsic hallucinations. These insights provide new directions for aligning detection strategies with the nature of hallucination and highlight attention as a rich signal for quantifying model uncertainty.
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。
