新型防御GARD可中和多模态模型后门攻击

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-11 05:00

研究人员开发了GARD，一种针对多模态对比学习模型（如CLIP）的后门攻击的新型防御机制。该方法，即梯度引导对抗性表示防御，通过使用对抗性探针来置换干净图像的表示，然后将其与分离的干净锚点重新对齐。GARD能有效抵御各种后门攻击，将成功率降至1%以下，同时保持高干净准确率，并且在推理时无需额外计算。 AI

影响增强了多模态AI模型抵御复杂后门攻击的安全性，可能增加在敏感应用中的信任度和采用率。

排序理由该集群包含一篇学术论文，详细介绍了一种防御AI模型免受特定类型攻击的新方法。[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: 32 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇学术论文，详细介绍了一种防御AI模型免受特定类型攻击的新方法。[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: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CV TIER_1 English(EN) · Yuhong Wang, Yujun Cai, Lei Feng, Songze Li · 2026-11-11 05:00

GARD：梯度引导对抗性表示防御多模态对比学习模型中的后门

arXiv:2611.14143v1 Announce Type: cross Abstract: Multimodal contrastive learning models such as CLIP are vulnerable to backdoor attacks with diverse trigger structures, ranging from localized patches to globally distributed perturbations, with recent attacks increasingly optimiz…

本文如何被排名

报道来源 [1]

GARD：梯度引导对抗性表示防御多模态对比学习模型中的后门

相关实体

相关话题