arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.14857cs.CLcs.AIcs.CR
收藏

系统消息对大语言模型中的越狱攻击真的重要吗?

Is the System Message Really Important to Jailbreaks in Large Language Models?

Xiaotian Zou, Yongkang Chen, Ke Li

AI总结:

该研究探究系统消息对大语言模型越狱攻击的影响,发现不同系统消息抗越狱能力有差异,提出SMEA算法生成高抗扰性系统消息,提升大模型安全性与越狱攻击门槛。

摘要:

大语言模型(LLMs)的快速发展使其成为现代社会不可或缺的工具。虽然通常会在发布前采取安全措施使LLMs与人类价值观对齐,但近期研究揭示了一个令人担忧的现象——“越狱(Jailbreak)”。该术语指LLMs在收到恶意问题提示时生成意外且潜在有害回复的情况。现有研究大多聚焦于越狱提示的生成,但实验中的系统消息配置差异极大。本文旨在回答一个问题:系统消息对LLMs中的越狱攻击真的重要吗?我们在主流LLMs中开展实验,使用不同系统消息(短消息、长消息、无消息)生成越狱提示,发现不同系统消息对越狱的抵抗能力存在差异。据此,我们探究了不同系统消息下越狱攻击在LLMs间的可迁移性。此外,我们提出系统消息进化算法(System Messages Evolutionary Algorithm, SMEA),用于生成对越狱提示抵抗性更强的系统消息,即便消息仅有微小改动也能生效。通过SMEA,我们得到了一组鲁棒的系统消息种群,且系统消息的长度几乎没有变化。本研究不仅增强了LLMs的安全性,也提升了越狱攻击的难度,推动了该研究领域的发展。

英文摘要:

The rapid evolution of Large Language Models (LLMs) has rendered them indispensable in modern society. While security measures are typically to align LLMs with human values prior to release, recent studies have unveiled a concerning phenomenon named "Jailbreak". This term refers to the unexpected and potentially harmful responses generated by LLMs when prompted with malicious questions. Most existing research focus on generating jailbreak prompts but system message configurations vary significantly in experiments. In this paper, we aim to answer a question: Is the system message really important for jailbreaks in LLMs? We conduct experiments in mainstream LLMs to generate jailbreak prompts with varying system messages: short, long, and none. We discover that different system messages have distinct resistances to jailbreaks. Therefore, we explore the transferability of jailbreaks across LLMs with different system messages. Furthermore, we propose the System Messages Evolutionary Algorithm (SMEA) to generate system messages that are more resistant to jailbreak prompts, even with minor changes. Through SMEA, we get a robust system messages population with little change in the length of system messages. Our research not only bolsters LLMs security but also raises the bar for jailbreaks, fostering advancements in this field of study.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑