Sign up to save your podcasts
Or
Share Jailbreaking with Universal Multi-Prompts
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
Jailbreaking with Universal Multi-Prompts
🤗 Upvotes: 4 | cs.CL, cs.AI, cs.CR, cs.LG
Authors:
Yu-Ling Hsu, Hsuan Su, Shang-Tse Chen
Title:
Jailbreaking with Universal Multi-Prompts
Arxiv:
http://arxiv.org/abs/2502.01154v1
Abstract:
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.
View all episodes
By Jingwen Liang, Gengyu Wang🤗 Upvotes: 4 | cs.CL, cs.AI, cs.CR, cs.LG
Authors:
Yu-Ling Hsu, Hsuan Su, Shang-Tse Chen
Title:
Jailbreaking with Universal Multi-Prompts
Arxiv:
http://arxiv.org/abs/2502.01154v1
Abstract:
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.