让检索更多样，提升复杂问答准确率
DF-RAG: Query-Aware Diversity for Retrieval-Augmented Generation
- 基于最大边际相关性，动态选择既相关又互异的文本片段
- 在复杂问答任务上，相比普通检索提升4%-10%的准确率
- 无需额外训练，适合需要高信息量的推理型问答场景
检索增强生成（RAG）常用于将语言模型输出与领域知识对齐，但在需要推理的问答任务中表现受限，因传统检索方法（如余弦相似度）仅追求相关性，导致内容冗余，降低信息召回。为此，本文提出多样性聚焦的检索增强生成（DF-RAG），在检索阶段系统引入多样性，提升复杂推理问答性能。DF-RAG基于最大边际相关性框架，筛选既与查询相关又彼此差异最大的信息块。其关键创新在于可动态调整每条查询的多样性水平，无需额外微调或先验信息。实验表明，DF-RAG在推理密集型问答基准上比纯余弦相似度检索提升4%-10%的F1值，优于多个现有基线。我们估计理想上限可达比原RAG高出18%的绝对F1提升，而DF-RAG已实现其中91.3%。
原文摘要 · Abstract (English)
Retrieval-augmented generation (RAG) is a common technique for grounding language model outputs in domain-specific information. However, RAG is often challenged by reasoning-intensive question-answering (QA), since common retrieval methods like cosine similarity maximize relevance at the cost of introducing redundant content, which can reduce information recall. To address this, we introduce Diversity-Focused Retrieval-Augmented Generation (DF-RAG), which systematically incorporates diversity into the retrieval step to improve performance on complex, reasoning-intensive QA benchmarks. DF-RAG builds upon the Maximal Marginal Relevance framework to select information chunks that are both relevant to the query and maximally dissimilar from each other. A key innovation of DF-RAG is its ability to optimize the level of diversity for each query dynamically at test time without requiring any additional fine-tuning or prior information. We show that DF-RAG improves F1 performance on reasoning-intensive QA benchmarks by 4-10 percent over vanilla RAG using cosine similarity and also outperforms other established baselines. Furthermore, we estimate an Oracle ceiling of up to 18 percent absolute F1 gains over vanilla RAG, of which DF-RAG captures up to 91.3 percent.
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。