PulseAugur
中
实时 22:57:33
English(EN) Poisoning Prompt-Guided Sampling in Video Large Language Models

新的PoisonVID攻击绕过了视频大语言模型的安全功能

作者 PulseAugur 编辑部 · [1 个来源] ·

研究人员开发了一种名为PoisonVID的新型攻击，可以绕过视频大语言模型（VideoLLMs）的安全措施。这些模型通过采样关键帧来审核用户生成的内容，但PoisonVID可以操纵这个采样过程。该攻击通过微妙地改变有害视频帧，使得VideoLLM的提示引导采样机制无法识别它们，从而有效地抑制安全警报。该方法在各种VideoLLM架构和有害内容类别中都显示出高成功率，甚至能够抵御多种防御机制。 AI

影响 这项研究揭示了当前视频大语言模型安全系统的一个关键漏洞，可能影响内容审核并需要新的防御策略。

排序理由 详细介绍针对AI模型新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 →

新的PoisonVID攻击绕过了视频大语言模型的安全功能

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
详细介绍针对AI模型新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
64 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]

  1. arXiv cs.CV TIER_1 English(EN) · Yuxin Cao, Wei Song, Jingling Xue, Jin Song Dong ·

    视频大语言模型中的中毒提示引导采样

    arXiv:2509.20851v2 Announce Type: replace Abstract: Video Large Language Models (VideoLLMs) are increasingly deployed as automated moderators on user-generated video platforms, where a few unwatched seconds of harmful footage are enough to suppress a safety alert. Because encodin…

相关实体

相关话题