LUMINA通过分析外部上下文与内部知识的使用差异，精准检测RAG系统中的幻觉。
LUMINA: Detecting Hallucinations in RAG System with Context-Knowledge Signals
- 用分布距离衡量外部上下文使用程度，通过层间预测词演变追踪内部知识依赖。
- 在HalluRAG上比现有方法最高提升13% AUROC，且不依赖高精度检索。
- 无需调参、抗干扰强，适合实际部署的RAG系统幻觉检测。
检索增强生成（RAG）通过引入外部文档来减少大语言模型（LLM）的幻觉，但即使提供正确充分的上下文，仍会出现幻觉。研究表明，这源于模型对外部上下文和内部知识的利用失衡。现有方法需大量超参数调优，泛化性差。本文提出LUMINA框架，通过量化外部上下文利用（基于分布距离）和内部知识利用（追踪跨Transformer层预测词演变）来检测幻觉，并引入统计验证机制。在常见RAG幻觉基准和四个开源LLM上测试，LUMINA表现稳定，最大提升13% AUROC（HalluRAG），且对检索质量要求宽松，具备实用性。
原文摘要 · Abstract (English)
Retrieval-Augmented Generation (RAG) aims to mitigate hallucinations in large language models (LLMs) by grounding responses in retrieved documents. Yet, RAG-based LLMs still hallucinate even when provided with correct and sufficient context. A growing line of work suggests that this stems from an imbalance between how models use external context and their internal knowledge, and several approaches have attempted to quantify these signals for hallucination detection. However, existing methods require extensive hyperparameter tuning, limiting their generalizability. We propose LUMINA, a novel framework that detects hallucinations in RAG systems through context--knowledge signals: external context utilization is quantified via distributional distance, while internal knowledge utilization is measured by tracking how predicted tokens evolve across transformer layers. We further introduce a framework for statistically validating these measurements. Experiments on common RAG hallucination benchmarks and four open-source LLMs show that LUMINA achieves consistently high AUROC and AUPRC scores, outperforming prior utilization-based methods by up to +13% AUROC on HalluRAG. Moreover, LUMINA remains robust under relaxed assumptions about retrieval quality and model matching, offering both effectiveness and practicality. LUMINA: https://github.com/deeplearning-wisc/LUMINA
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。