来源：arXiv cs.AI · 2026-10-05 · 新型 AI / 新品发布

arXiv:2610.02300v1 Announce Type: new Abstract: Training-free safeguards for text-to-image generation often rely on a reusable safety signal, such as an unsafe direction or global toxic subspace, appl

查看原文 →

本条资讯由「小柚晨报」整理自 arXiv cs.AI，仅作交流参考，版权归原作者所有。