TrojanRAG: Retrieval-Augmented Generation Can Be Backdoor Driver in Large Language Models

arXiv:2405.13401v1 Announce Type: new
Abstract: Large language models (LLMs) have raised concerns about potential security threats despite performing significantly in Natural Language Processing (NLP). Backdoor attacks initially verified that LLM is doing substantial harm at all stages, but the cost and robustness have been criticized. Attacking LLMs is inherently risky in security review, while prohibitively expensive. Besides, the continuous iteration of LLMs will degrade the robustness of backdoors. In this paper, we propose TrojanRAG, which employs a joint backdoor attack …

arxiv attacks backdoor backdoor attacks can cost cs.cl cs.cr doing driver harm language language models large llm llms natural natural language natural language processing nlp performing robustness security security threats threats verified