RAG 三重稳健性分析：揭示通用病理与条件后果
原标题：Universal Pathologies, Conditional Consequences: A Triple-Robustness Analysis of RAG for Multi-Hop Traceability
AI 摘要
一项针对检索增强生成（RAG）架构的三重稳健性分析发现，GraphRAG 的过度引用问题在架构上普遍存在，但其对忠实度的影响取决于语料库。在 DO-178C 类型化边需求上，GraphRAG 的忠实度随跳数增加而下降，而在 Wikipedia 链上则上升。研究还发现，单一 LLM 评判器的忠实度评估对检索状态敏感，而基于稠密嵌入的路由器在跳数分类上达到 0.86 的宏 F1。作者认为三重稳健性分析是可信 RAG 架构声明的最低标准。
正文节选
Computer Science > Computation and Language Title:Universal Pathologies, Conditional Consequences: A Triple-Robustness Analysis of RAG for Multi-Hop Traceability View PDF HTML (experimental) Abstract:GraphRAG underperforms vector RAG on citation precision in many reports, but where and why have remained corpus-bound. We present a triple-robustness analysis that holds the retrieval architecture fixed and varies three orthogonal axes embedder (local e5-small -> Azure text-embedding-3-s