Retrieval Augmented Generation (RAG) frameworks improve the accuracy of large language models (LLMs) by integrating external knowledge from retrieved documents, thereby overcoming the limitations of models’ static intrinsic knowledge. However, these systems are susceptible to adversarial attacks that manipulate the retrieval process by introducing documents that are adversarial yet semantically similar to the query.

We propose GRADA, a simple yet effective Graph-based Reranking against Adversarial Document Attacks framework aiming at preserving retrieval quality while significantly reducing the success of adversaries. Our experiments on five LLMs demonstrate up to an 80% reduction in attack success rates while maintaining minimal loss in accuracy.