Certifiably Robust RAG against Retrieval Corruption

Understand

Retrieval-augmented generation (RAG) is susceptible to retrieval corruption attacks, where malicious passages injected into retrieval results can lead to inaccurate model responses.

• We propose RobustRAG, the first defense framework with certifiable robustness against retrieval corruption attacks.
• The key insight of RobustRAG is an isolate-then-aggregate strategy: we isolate passages into disjoint groups, generate LLM responses based on the concatenated passages from each isolated group, and then securely aggregate these responses for a robust output.
• To instantiate RobustRAG, we design keyword-based and decoding-based algorithms for securely aggregating unstructured text responses.