BadEdit: Backdooring large language models by model editing

arXiv:2403.13355v1 Announce Type: new
Abstract: Mainstream backdoor attack methods typically demand substantial tuning data for poisoning, limiting their practicality and potentially degrading the overall performance when applied to Large Language Models (LLMs). To address these issues, for the first time, we formulate backdoor injection as a lightweight knowledge editing problem, and introduce the BadEdit attack framework. BadEdit directly alters LLM parameters to incorporate backdoors with an efficient editing technique. It boasts superiority over existing backdoor injection techniques in several areas: …

address arxiv attack backdoor backdoor attack backdooring cs.ai cs.cr data demand editing injection knowledge language language models large llms mainstream performance poisoning problem