English(EN) A Cross-Modal Prompt Injection Attack against Large Vision-Language Models with Image-Only Perturbation

新型跨模态提示注入攻击目标锁定大型视觉语言模型

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-27 04:00

研究人员开发了一种名为CrossMPI的新型跨模态提示注入攻击，该攻击仅通过图像扰动即可操纵大型视觉语言模型（LVLMs）对文本和图像输入的解释。该攻击在模型隐藏状态空间内优化扰动，该空间比典型的视觉嵌入空间具有更多的参数。CrossMPI采用层选择等策略来识别多模态集成中的关键层，并采用距离递减预算分配的图像扰动来提高其有效性。 AI

影响这项研究揭示了LVLMs中的一种新漏洞，可能影响多模态AI系统的安全性和可靠性。

排序理由详细介绍针对AI模型的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: 6 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 详细介绍针对AI模型的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: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CV TIER_1 English(EN) · Hao Yang, Zhuo Ma, Yang Liu, Yilong Yang, Guancheng Wang, JianFeng Ma · 2026-10-27 04:00

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

arXiv:2605.16090v2 Announce Type: replace-cross Abstract: 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…

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