Mitigating Fine-tuning Jailbreak Attack with Backdoor Enhanced Alignment

arXiv:2402.14968v1 Announce Type: cross
Abstract: Despite the general capabilities of Large Language Models (LLMs) like GPT-4 and Llama-2, these models still request fine-tuning or adaptation with customized data when it comes to meeting the specific business demands and intricacies of tailored use cases. However, this process inevitably introduces new safety threats, particularly against the Fine-tuning based Jailbreak Attack (FJAttack), where incorporating just a few harmful examples into the fine-tuning dataset can significantly compromise the model safety. Though potential defenses have …

abstract alignment arxiv backdoor business capabilities cases cs.cl cs.cr data fine-tuning general gpt gpt-4 jailbreak language language models large language large language models llama llms process safety threats type use cases