PRA-RAG: Provably Robust Aggregation in Retrieval-Augmented Generation against Retrieval Corruption
For RAG systems vulnerable to retrieval corruption, this work provides the first provable robustness guarantees with strong empirical performance.
PRA-RAG proposes a provably robust aggregation algorithm for RAG that defends against poisoning attacks on retrieved texts, reducing attack success rate to as low as 1% while maintaining 71% accuracy, significantly outperforming existing methods.
Retrieval-Augmented Generation (RAG) enhances Large Language Models (LLMs) by incorporating external knowledge, effectively mitigating their inherent knowledge limitations. However, RAG remains vulnerable to poisoning attacks that manipulate retrieved texts to mislead model outputs. Existing defense mechanisms often lack theoretical robustness guarantees and perform unreliably when the LLM has limited knowledge of the retrieved content. In this work, we propose PRA-RAG, a provably robust retrieval aggregation algorithm designed to defend against poisoning attacks on retrieved texts. PRA-RAG samples multiple combinations of retrieved texts and utilizes geometric structures in the embedding space to identify a robust subset, from which a stable aggregated representation is derived. We provide theoretical bounds on the maximum impact of poisoned retrieved content and establish a quantitative measure of RAG's robustness. Experiments across multiple benchmarks and RAG architectures demonstrate that PRA-RAG reduces the attack success rate to as low as 1% while maintaining an accuracy of 71%, significantly outperforming representative state-of-the-art methods.