文档越多，检索越差？用领域分块解决向量搜索稀释问题
When More Documents Hurt RAG: Mitigating Vector Search Dilution with Domain-Scoped, Model-Agnostic Retrieval
- 按组织元数据对文档分域，提升检索精准度
- 在5个语料库上，P@10从0.77提升至0.86（显著提升）
- 适合大规模异构文档的RAG系统，尤其关注准确率的场景
当检索增强生成系统扩展到大规模、异构文档集合时，密集相似度的区分能力下降，Top-k检索频繁返回语义相近但上下文错误的文本片段，这种现象称为向量搜索稀释。我们在怀俄明州交通部的部署中观察到，文档数从54增至1,128（共88,907个片段）后，准确率由75%降至40%以下。为此提出MASDR-RAG：基于组织元数据的多智能体领域分块检索。在200个专家验证查询上，针对五种大模型、六种语料库和两种索引结构进行评估，结果表明领域分块可显著提升P@10至0.86（p < 0.05）。多智能体编排研究揭示精度与忠实性间的矛盾，建议优先分域，再单次合成；仅在真正跨领域且支持原生工具调用的模型中使用复杂编排。代码与数据将在接受后公开。
原文摘要 · Abstract (English)
Retrieval-augmented generation degrades when scaled to large, heterogeneous document collections, where dense similarity loses discriminative power, and top-k retrieval increasingly returns semantically similar but contextually incorrect chunks. We refer to this failure mode as vector search dilution. Even when using hybrid dense+sparse retrieval, we observed this firsthand in a deployed Wyoming Department of Transportation corpus, where scaling from 54 to 1,128 documents (88,907 chunks) reduced accuracy from 75% to below 40%. To address this dilution, we propose MASDR-RAG ( Multi-Agent Scoped Domain Retrieval for RAG) and evaluate it on 200 expert-validated queries across five LLM backbones, six corpora, and two index stacks. Our results indicate that domain scoping using organizational metadata is the key fix, significantly improving P@10 from 0.77 to 0.86 ($p < 0.05$). Furthermore, our investigation of multi-agent orchestration revealed that a high degree of configuration dependence results --creating what we call the precision-faithfulness paradox. Based on these varied outcomes, our practical recommendation is simple: scope first, then perform a single synthesis call, reserving full multi-agent orchestration for genuinely multi-domain corpora paired with native-tool-call backbones. Code and Data will be made public upon acceptance.
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。