TombRaider:进入历史宝库以越狱大语言模型
TombRaider: Entering the Vault of History to Jailbreak Large Language Models
AI总结:
针对大语言模型(LLM)应用的越狱攻击问题,提出利用LLM历史知识能力的TombRaider技术,通过检查智能体提取历史信息、攻击智能体生成对抗提示,在实验中表现优于现有技术,凸显LLM安全防护漏洞并强调需更 robust 防御。
摘要:
警告:本文包含可能涉及潜在有害行为的内容,仅为研究目的进行严格讨论。越狱攻击会阻碍大语言模型(LLM)应用的安全性,尤其是聊天机器人。研究越狱技术是提升这些应用安全性的重要AI红队任务。本文介绍TombRaider,一种利用LLM存储、检索和使用历史知识能力的新型越狱技术。TombRaider采用两个智能体:检查智能体提取相关历史信息,攻击智能体生成对抗性提示,从而有效绕过安全过滤器。我们在六个流行模型上对TombRaider进行了密集评估,实验结果显示其性能优于最先进的越狱技术,在裸模型上实现近100%的攻击成功率(ASR),面对防御机制时仍保持55.4%以上的ASR。我们的发现凸显了现有LLM安全防护的关键漏洞,强调了对更 robust 安全防御的需求。
英文摘要:
Warning: This paper contains content that may involve potentially harmful behaviours, discussed strictly for research purposes. Jailbreak attacks can hinder the safety of Large Language Model (LLM) applications, especially chatbots. Studying jailbreak techniques is an important AI red teaming task for improving the safety of these applications. In this paper, we introduce TombRaider, a novel jailbreak technique that exploits the ability to store, retrieve, and use historical knowledge of LLMs. TombRaider employs two agents, the inspector agent to extract relevant historical information and the attacker agent to generate adversarial prompts, enabling effective bypassing of safety filters. We intensively evaluated TombRaider on six popular models. Experimental results showed that TombRaider could outperform state-of-the-art jailbreak techniques, achieving nearly 100% attack success rates (ASRs) on bare models and maintaining over 55.4% ASR against defence mechanisms. Our findings highlight critical vulnerabilities in existing LLM safeguards, underscoring the need for more robust safety defences.
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。