新型攻击通过仅图像提示操纵视觉语言模型

作者 PulseAugur 编辑部 · [1 个来源] · 2026-05-15 15:47

研究人员开发了一种名为CrossMPI的新型跨模态提示注入攻击，该攻击可以通过仅图像的扰动来操纵大型视觉语言模型（LVLMs）对文本和图像输入的解释。该攻击针对模型的隐藏状态空间，该空间比以往攻击中通常使用的视觉嵌入空间更大且更具影响力。该方法采用层选择策略来关注关键的多模态集成层，并采用距离递减预算分配来约束图像扰动，在实验中表现优于现有方法。 AI

影响这项研究揭示了视觉语言模型中的一种新漏洞，可能影响其安全部署并需要新的防御机制。

排序理由该集群包含一篇详细介绍新型攻击方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍新型攻击方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, safety; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 165 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CV TIER_1 English(EN) · JianFeng Ma · 2026-05-15 15:47

针对大型视觉语言模型图像唯一扰动的跨模态提示注入攻击

Large vision-language models (LVLMs) have emerged as a powerful paradigm for multimodal intelligence, but their growing deployment also expands the attack surface of prompt injection. Despite this growing concern, existing attacks still suffer from a critical limitation: the inje…

本文如何被排名

报道来源 [1]

针对大型视觉语言模型图像唯一扰动的跨模态提示注入攻击

相关实体

相关话题