研究人员开发了一个名为TooBad的新后门攻击框架，专门针对扩散模型。该框架通过采用针对扩散模型量身定制的新颖触发器优化技术，显著提高了后门攻击的性能。TooBad在非常低的中毒率（0.5%）和最少的训练周期下展示了高攻击成功率，使其隐蔽且高效，同时能规避当前最先进的防御措施。 AI
影响 突出了扩散模型中的关键漏洞，有必要开发更强大的防御措施来抵御隐蔽且高效的攻击。
排序理由 详细介绍扩散模型后门攻击新方法的学术论文。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2606.26285v1 Announce Type: cross Abstract: Noise-based backdoor attacks on diffusion models typically rely on input-time trigger injection, untargeted activation, and out-of-distribution target generation. Such assumptions reduce both the stealthiness and the practical rel…
Diffusion models (DMs), despite their impressive capabilities across a wide range of generative tasks, have been shown to be vulnerable to backdoor attacks. However, existing backdoor methods face critical trade-offs among key factors: attack performance, stealthiness, time compl…
Diffusion models (DMs), despite their impressive capabilities across a wide range of generative tasks, have been shown to be vulnerable to backdoor attacks. However, existing backdoor methods face critical trade-offs among key factors: attack performance, stealthiness, time compl…