Skip to content
Search
paperMay 2026Unreviewed

Architecture Matters: Comparing RAG Systems under Knowledge Base Poisoning

Samuel Korn

View resourcePDF

Abstract

Retrieval-Augmented Generation (RAG) systems are vulnerable to knowledge base poisoning, yet existing attacks have been evaluated almost exclusively against vanilla retrieve-then-generate pipelines. Architectures designed to handle conflicting retrieved information - multi-agent debate, agentic retrieval, recursive language models - remain untested against adversarially optimized contradictions. We evaluate four RAG architectures (vanilla RAG, agentic RAG, MADAM-RAG, and Recursive Language Model

Categories

Agentic ThreatsAgent Architecture

Cite

@misc{korn2026architecture,
  title = {{Architecture Matters: Comparing RAG Systems under Knowledge Base Poisoning}},
  author = {Samuel Korn},
  year = {2026},
  month = may,
  eprint = {2605.05632},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2605.05632}
}