paper-with-me

홈 › Papers

When More Documents Hurt RAG: Mitigating Vector Search Dilution with Domain-Scoped, Model-Agnostic Retrieval

2026-06-09 · Nabaraj Subedi, Ahmed Abdelaty, Shivanand Venkanna Sheshappanavar arxiv

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.

📄 PDF Abstract BibTeX arXiv:2606.11350

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

SOLAR: Sparse Orthogonal Learned and Random Embeddings

2020-08-30 · ICLR 2021 1 · Tharun Medini, Beidi Chen, Anshumali Shrivastava

Dense embedding models are commonly deployed in commercial search engines, wherein all the document vectors are pre-computed, and near-neighbor search (NNS) is performed with the query vector to find relevant documents. …

Multi-Label ClassificationMUlTI-LABEL-ClASSIFICATION

Vietnamese Legal Information Retrieval in Question-Answering System

2024-09-05 · Thiem Nguyen Ba, Vinh Doan The, Tung Pham Quang, Toan Tran Van

In the modern era of rapidly increasing data volumes, accurately retrieving and recommending relevant documents has become crucial in enhancing the reliability of Question Answering (QA) systems. Recently, Retrieval Augm…

HallucinationInformation RetrievalQuestion AnsweringRAG+3

Progressive Searching for Retrieval in RAG

2026-02-07 · Taehee Jeong, Xingzhe Zhao, Peizu Li, Markus Valvur 외 arxiv

Retrieval Augmented Generation (RAG) is a promising technique for mitigating two key limitations of large language models (LLMs): outdated information and hallucinations. RAG system stores documents as embedding vectors …

The Curious Case of Adversarially Robust Models: More Data Can Help, Double Descend, or Hurt Generalization

2020-02-25 · Yifei Min, Lin Chen, Amin Karbasi

Adversarial training has shown its ability in producing models that are robust to perturbations on the input data, but usually at the expense of decrease in the standard accuracy. To mitigate this issue, it is commonly b…

ClassificationGeneral Classification

Comparing Lexical and Semantic Vector Search Methods When Classifying Medical Documents

2025-05-16 · Lee Harris, Philippe De Wilde, James Bentham

Classification is a common AI problem, and vector search is a typical solution. This transforms a given body of text into a numerical representation, known as an embedding, and modern improvements to vector search focus …

Information Retrieval