English(EN) ShieldCLIP: Selective Safety Alignment for Harmful Content Mitigation in Multimodal Foundation Models

ShieldCLIP框架增强多模态模型的安全对齐

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-01 04:00

研究人员推出ShieldCLIP，一个旨在通过选择性处理有害内容来增强多模态基础模型安全对齐的新型框架。与以往将所有生成样本都视为不安全的先前方法不同，ShieldCLIP根据单个模态的安全状态进行区分。该方法得到了ViSUv2的支持，这是一个具有按模态安全标签的新数据集。ShieldCLIP已证明在减少各种任务中的有害输出方面是有效的，包括使用Stable Diffusion v1.4和SDXL等模型的跨模态检索和文本到图像生成，同时保留了原始嵌入空间的效用。 AI

影响这项研究可能导致多模态AI系统中更强大的安全机制，在不损害良性内容的情况下减少有害输出。

排序理由该集群描述了一篇介绍AI安全新框架和数据集的新研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 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: 22 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群描述了一篇介绍AI安全新框架和数据集的新研究论文。[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: 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.AI TIER_1 English(EN) · Tobia Poppi, Silvia Cappelletti, Samuele Poppi, Marcella Cornia, Lorenzo Baraldi, Diego Garcia-Olano, Rita Cucchiara · 2026-10-01 04:00

ShieldCLIP：多模态基础模型有害内容缓解的选择性安全对齐

arXiv:2609.39688v1 Announce Type: cross Abstract: Multimodal encoders such as CLIP underlie many downstream systems, but their web-scale training data embed harmful associations that safety alignment must suppress without unnecessarily changing benign representations. Because eth…

本文如何被排名

报道来源 [1]

ShieldCLIP：多模态基础模型有害内容缓解的选择性安全对齐

相关实体

相关话题