arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.15050cs.LGcs.AIcs.CRcs.MM
收藏

Arondight:利用自动生成的多模态越狱提示对大型视觉语言模型进行红队测试

Arondight: Red Teaming Large Vision Language Models with Auto-generated Multi-modal Jailbreak Prompts

City University of Hong Kong(香港城市大学);City University of Hong Kong (Dongguan)(香港城市大学(东莞));University of Science and Technology Beijing(北京科技大学);The University of Melbourne(墨尔本大学)

Yi Liu, Chengjun Cai, Xiaoli Zhang, Xingliang Yuan, Cong Wang

AI总结:

针对VLM缺乏红队框架的问题,提出Arondight,利用红队VLM和强化学习引导的LLM自动生成多模态越狱提示,在GPT-4上达到84.5%平均攻击成功率,暴露安全漏洞并提供加固建议。

摘要:

大型视觉语言模型(VLM)扩展并增强了大型语言模型(LLM)的感知能力。尽管这些进展为LLM应用提供了新的可能性,但也引发了重大的安全和伦理问题,尤其是在生成有害内容方面。虽然LLM借助红队测试框架已经进行了广泛的安全评估,但VLM目前缺乏一个完善的红队测试框架。为了填补这一空白,我们引入了Arondight,一个专门为VLM定制的标准化红队框架。Arondight致力于解决将现有红队测试方法从LLM迁移到VLM时遇到的视觉模态缺失和多样性不足的问题。我们的框架具有自动化的多模态越狱攻击功能,其中视觉越狱提示由红队VLM生成,文本提示由受强化学习智能体引导的红队LLM生成。为了增强VLM安全评估的全面性,我们整合了熵奖励和新颖性奖励指标。这些元素激励强化学习智能体引导红队LLM生成更广泛、多样且前所未有的测试用例。我们对十个前沿VLM的评估暴露了显著的安全漏洞,特别是在生成有毒图像和对齐多模态提示方面。具体而言,我们的Arondight在OpenAI定义的所有十四个禁止场景中,在生成有毒文本方面,对GPT-4实现了84.5%的平均攻击成功率。为了更清晰的比较,我们还根据安全级别对现有VLM进行了分类,并提供了相应的强化建议。我们的多模态提示数据集和红队代码将在伦理委员会批准后发布。内容警告:本文包含有害的模型响应。

英文摘要:

Large Vision Language Models (VLMs) extend and enhance the perceptual abilities of Large Language Models (LLMs). Despite offering new possibilities for LLM applications, these advancements raise significant security and ethical concerns, particularly regarding the generation of harmful content. While LLMs have undergone extensive security evaluations with the aid of red teaming frameworks, VLMs currently lack a well-developed one. To fill this gap, we introduce Arondight, a standardized red team framework tailored specifically for VLMs. Arondight is dedicated to resolving issues related to the absence of visual modality and inadequate diversity encountered when transitioning existing red teaming methodologies from LLMs to VLMs. Our framework features an automated multi-modal jailbreak attack, wherein visual jailbreak prompts are produced by a red team VLM, and textual prompts are generated by a red team LLM guided by a reinforcement learning agent. To enhance the comprehensiveness of VLM security evaluation, we integrate entropy bonuses and novelty reward metrics. These elements incentivize the RL agent to guide the red team LLM in creating a wider array of diverse and previously unseen test cases. Our evaluation of ten cutting-edge VLMs exposes significant security vulnerabilities, particularly in generating toxic images and aligning multi-modal prompts. In particular, our Arondight achieves an average attack success rate of 84.5\% on GPT-4 in all fourteen prohibited scenarios defined by OpenAI in terms of generating toxic text. For a clearer comparison, we also categorize existing VLMs based on their safety levels and provide corresponding reinforcement recommendations. Our multimodal prompt dataset and red team code will be released after ethics committee approval. CONTENT WARNING: THIS PAPER CONTAINS HARMFUL MODEL RESPONSES.

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。

↑