ADMIT: Few-shot Knowledge Poisoning Attacks on RAG-based Fact Checking

ArXi:2510.13842v2 Announce Type: replace-cross Knowledge poisoning poses a critical threat to Retrieval-Augmented Generation (RAG) systems by injecting adversarial content into knowledge bases, tricking Large Language Models (LLMs) into producing attacker-controlled outputs grounded in manipulated context. Prior work highlights LLMs' susceptibility to misleading or malicious retrieved content. However, real-world fact-checking scenarios are challenging, as credible evidence typically dominates the retrieval pool.