1 paper
Xiaonan Si, Meilin Zhu, Simeng Qin +7
Retrieval-augmented generation (RAG) systems enhance large language models (LLMs) with external knowledge but are vulnerable to corpus poisoning and contamination attacks, which ca…