面向证据不确定性和幻觉的面级追踪（RAG）
Facet-Level Tracing of Evidence Uncertainty and Hallucination in RAG
AI总结：
本文提出了一种面向问答任务的面级诊断框架，通过分解每个输入问题为原子推理面，评估证据充分性和基础性，揭示RAG系统中幻觉产生的根源在于生成过程中证据整合的方式，而非检索准确性。
摘要：
检索增强生成（RAG）旨在通过检索证据来减少幻觉，但即使有相关文档可用，幻觉答案仍然常见。现有评估集中在答案级或段落级准确性，对生成过程中证据使用缺乏深入理解。本文引入了一种面向问答任务的面级诊断框架，将每个输入问题分解为原子推理面。对于每个面，我们使用结构化的Facet x Chunk矩阵评估证据充分性和基础性，该矩阵结合检索相关性和基于自然语言推理的忠实度评分。为了诊断证据使用，我们分析了三种受控推理模式：严格RAG，强制依赖检索证据；软RAG，允许整合检索证据和参数知识；以及无检索的LLM生成。比较这些模式能够深入分析检索生成不一致，即相关证据被检索但未在生成过程中正确整合的情况。在医疗问答和HotpotQA上，我们评估了三种开源和闭源LLM（GPT、Gemini和LLaMA），提供了可解释的诊断，揭示了反复出现的面级失败模式，包括证据缺失、证据不一致和先验驱动覆盖。我们的结果表明，RAG系统中的幻觉更多由生成过程中证据整合的方式决定，而非检索准确性，面级分析揭示了系统性的证据覆盖和不一致模式，这些模式在答案级评估下是隐藏的。
英文摘要：
Retrieval-Augmented Generation (RAG) aims to reduce hallucination by grounding answers in retrieved evidence, yet hallucinated answers remain common even when relevant documents are available. Existing evaluations focus on answer-level or passage-level accuracy, offering limited insight into how evidence is used during generation. In this work, we introduce a facet-level diagnostics framework for QA that decomposes each input question into atomic reasoning facets. For each facet, we assess evidence sufficiency and grounding using a structured Facet x Chunk matrix that combines retrieval relevance with natural language inference-based faithfulness scores. To diagnose evidence usage, we analyze three controlled inference modes: Strict RAG, which enforces exclusive reliance on retrieved evidence; Soft RAG, which allows integration of retrieved evidence and parametric knowledge; and LLM-only generation without retrieval. Comparing these modes enables thorough analysis of retrieval-generation misalignment, defined as cases where relevant evidence is retrieved but not correctly integrated during generation. Across medical QA and HotpotQA, we evaluate three open-source and closed-source LLMs (GPT, Gemini, and LLaMA), providing interpretable diagnostics that reveal recurring facet-level failure modes, including evidence absence, evidence misalignment, and prior-driven overrides. Our results demonstrate that hallucinations in RAG systems are driven less by retrieval accuracy and more by how retrieved evidence is integrated during generation, with facet-level analysis exposing systematic evidence override and misalignment patterns that remain hidden under answer-level evaluation.
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。
