RevPRAG: Revealing Poisoning Attacks in Retrieval-Augmented Generation through LLM Activation Analysis

TL;DR

RevPRAG is a novel detection method that uses LLM activation patterns to identify poisoning attacks in retrieval-augmented generation systems, significantly improving detection accuracy.

Contribution

This work introduces RevPRAG, the first automated detection pipeline leveraging LLM activations to identify poisoned responses in RAG systems.

Findings

Achieves 98% true positive rate in detecting poisoned responses

Maintains false positive rate close to 1%

Effective across multiple datasets and RAG architectures

Abstract

Retrieval-Augmented Generation (RAG) enriches the input to LLMs by retrieving information from the relevant knowledge database, enabling them to produce responses that are more accurate and contextually appropriate. It is worth noting that the knowledge database, being sourced from publicly available channels such as Wikipedia, inevitably introduces a new attack surface. RAG poisoning involves injecting malicious texts into the knowledge database, ultimately leading to the generation of the attacker's target response (also called poisoned response). However, there are currently limited methods available for detecting such poisoning attacks. We aim to bridge the gap in this work. Particularly, we introduce RevPRAG, a flexible and automated detection pipeline that leverages the activations of LLMs for poisoned response detection. Our investigation uncovers distinct patterns in LLMs' activations when generating correct responses versus poisoned responses. Our results on multiple benchmark datasets and RAG architectures show our approach could achieve 98% true positive rate, while maintaining false positive rates close to 1%.