arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.17262cs.CLcs.AI
收藏

不合时宜的发言:大语言模型在多轮对话中的安全漏洞

Speak Out of Turn: Safety Vulnerability of Large Language Models in Multi-turn Dialogue

Beijing University of Posts and Telecommunications(北京邮电大学);University of Michigan(密歇根大学)

Zhenhong Zhou, Jiuyang Xiang, Haopeng Chen, Quan Liu, Zherui Li, Sen Su

AI总结:

该研究聚焦大语言模型多轮对话的安全盲区,提出通过拆解有害查询为多轮子问题的越狱方法,实验验证多款LLM存在相关漏洞,为LLM安全防护带来新挑战。

摘要:

大语言模型(LLM)已被证实会生成非法或不道德的回复,尤其是在遭遇“越狱(jailbreak)”攻击时。越狱相关研究已凸显出LLM的安全问题,但现有研究大多聚焦单轮对话场景,忽略了多轮对话这一人类从LLM获取信息的核心模式所潜藏的复杂性与风险。本文提出,人类可利用多轮对话诱导LLM生成有害信息:即便多轮对话中的每一轮都紧密服务于同一个恶意目的,LLM可能也不会拒绝带有警示性或处于安全边界的模糊查询。据此,我们通过将不安全查询拆解为多个子查询开展多轮对话,诱导LLM逐步回答有害子问题,最终生成整体有害的回复。我们在多款LLM上开展的实验表明,当前LLM的安全机制在多轮对话场景中存在不足。本研究揭示了LLM在多轮对话复杂场景下的漏洞,为LLM安全研究提出了新挑战。

英文摘要:

Large Language Models (LLMs) have been demonstrated to generate illegal or unethical responses, particularly when subjected to "jailbreak." Research on jailbreak has highlighted the safety issues of LLMs. However, prior studies have predominantly focused on single-turn dialogue, ignoring the potential complexities and risks presented by multi-turn dialogue, a crucial mode through which humans derive information from LLMs. In this paper, we argue that humans could exploit multi-turn dialogue to induce LLMs into generating harmful information. LLMs may not intend to reject cautionary or borderline unsafe queries, even if each turn is closely served for one malicious purpose in a multi-turn dialogue. Therefore, by decomposing an unsafe query into several sub-queries for multi-turn dialogue, we induced LLMs to answer harmful sub-questions incrementally, culminating in an overall harmful response. Our experiments, conducted across a wide range of LLMs, indicate current inadequacies in the safety mechanisms of LLMs in multi-turn dialogue. Our findings expose vulnerabilities of LLMs in complex scenarios involving multi-turn dialogue, presenting new challenges for the safety of LLMs.

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。

↑