Jailbreaking with Universal Multi-Prompts
This addresses security vulnerabilities in LLMs for AI safety applications, representing an incremental advance in adversarial attack methods.
The paper tackles the problem of jailbreaking large language models by introducing JUMP, a prompt-based method using universal multi-prompts that can transfer to unseen tasks, and demonstrates it outperforms existing techniques in experiments.
Large language models (LLMs) have seen rapid development in recent years, revolutionizing various applications and significantly enhancing convenience and productivity. However, alongside their impressive capabilities, ethical concerns and new types of attacks, such as jailbreaking, have emerged. While most prompting techniques focus on optimizing adversarial inputs for individual cases, resulting in higher computational costs when dealing with large datasets. Less research has addressed the more general setting of training a universal attacker that can transfer to unseen tasks. In this paper, we introduce JUMP, a prompt-based method designed to jailbreak LLMs using universal multi-prompts. We also adapt our approach for defense, which we term DUMP. Experimental results demonstrate that our method for optimizing universal multi-prompts outperforms existing techniques.