Jiale Ding

RedTopic: Toward Topic-Diverse Red Teaming of Large Language Models

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