paperJuly 2026Unreviewed

RAGuard: A Layered Defense Framework for Retrieval-Augmented Generation Systems Against Data Poisoning

Pushkal Kumar, Tucker Nielson, Tanish Kolhe, Shubham Zala, Vincent Li

Abstract

Retrieval-Augmented Generation (RAG) systems ground large language models (LLMs) in external corpora, but this reliance exposes them to corpus poisoning: maliciously injected passages that manipulate retrieved evidence. We introduce RAGuard, a layered defense against \emph{factual} corpus-poisoning attacks on RAG pipelines. The first layer adversarially fine-tunes a dense retriever on synthetic poisoned documents (fabricated facts, contradictions, and reasoning traps), teaching it to downrank ma

Framework mappings

OWASP Top 10 for LLM Applications

LLM04Data and Model Poisoning

MITRE ATLAS

AML.T0020Poison Training Data

Suggested from the entry's categories.

Cite

@misc{kumar2026raguard,
  title = {{RAGuard: A Layered Defense Framework for Retrieval-Augmented Generation Systems Against Data Poisoning}},
  author = {Pushkal Kumar and Tucker Nielson and Tanish Kolhe and Shubham Zala and Vincent Li},
  year = {2026},
  month = jul,
  eprint = {2607.26339},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2607.26339}
}