Skip to content
Search
paperFebruary 2024ReviewedOpen access

PoisonedRAG: Knowledge Poisoning Attacks to Retrieval-Augmented Generation of Large Language Models

Wei Zou, Runpeng Geng, Binghui Wang, Jinyuan Jia

arXiv preprint

View resourcePDF

Abstract

Demonstrates knowledge poisoning attacks against RAG systems where adversaries inject malicious texts into the knowledge database to manipulate LLM outputs.

Categories

Data PoisoningRAG Security
#rag#knowledge-poisoning#vector-database

Framework mappings

OWASP Top 10 for LLM Applications
  • LLM04Data and Model Poisoning
  • LLM08Vector and Embedding Weaknesses
MITRE ATLAS
  • AML.T0020Poison Training Data

Cite

@misc{zou2024poisonedrag,
  title = {{PoisonedRAG: Knowledge Poisoning Attacks to Retrieval-Augmented Generation of Large Language Models}},
  author = {Wei Zou and Runpeng Geng and Binghui Wang and Jinyuan Jia},
  year = {2024},
  month = feb,
  eprint = {2402.07867},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2402.07867}
}