评测大模型识别信息权威性的能力，提升检索增强生成的可靠性
AuthorityBench: Benchmarking LLM Authority Perception for Reliable Retrieval-Augmented Generation
- 构建三类权威性数据集，评估模型对信息源可信度的判断力
- 发现结合网页内容反而降低判断准确率，权威性与文本风格无关
- 权威性过滤可显著提升RAG生成答案准确率，适合可信AI系统研发者
检索增强生成（RAG）虽能拓展大模型的外部知识，但仍易受低权威信息源误导。本文研究大模型是否具备感知信息权威性的能力，超越语义理解层面。为此提出AuthorityBench基准，包含三个数据集：DomainAuth（10,000个网页域名，基于PageRank的权威性评分）、EntityAuth（22,000个实体，基于流行度的权威性评分）和RAGAuth（120个查询，对应不同权威性的文档用于下游评估）。采用三种判别方法（PointJudge、PairJudge、ListJudge）及多种输出格式，评估五种LLM。结果显示，ListJudge与PairJudge搭配PointScore输出与真实权威性相关性最强，且ListJudge成本效益最优。值得注意的是，引入网页正文文本会持续降低判断性能，表明权威性并非文本风格的体现。下游RAG实验表明，基于权威性的过滤机制显著提升回答准确性，验证了该能力在可靠知识检索中的实际价值。代码与数据集已开源。
原文摘要 · Abstract (English)
Retrieval-Augmented Generation (RAG) enhances Large Language Models (LLMs) with external knowledge but remains vulnerable to low-authority sources that can propagate misinformation. We investigate whether LLMs can perceive information authority - a capability extending beyond semantic understanding. To address this, we introduce AuthorityBench, a comprehensive benchmark for evaluating LLM authority perception comprising three datasets: DomainAuth (10K web domains with PageRank-based authority), EntityAuth (22K entities with popularity-based authority), and RAGAuth (120 queries with documents of varying authority for downstream evaluation). We evaluate five LLMs using three judging methods (PointJudge, PairJudge, ListJudge) across multiple output formats. Results show that ListJudge and PairJudge with PointScore output achieve the strongest correlation with ground-truth authority, while ListJudge offers optimal cost-effectiveness. Notably, incorporating webpage text consistently degrades judgment performance, suggesting authority is distinct from textual style. Downstream experiments on RAG demonstrate that authority-guided filtering largely improves answer accuracy, validating the practical importance of authority perception for reliable knowledge retrieval. Code and benchmark are available at: https://github.com/Trustworthy-Information-Access/AuthorityBench.
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。