AdvFD：通过对抗 Fréchet 距离损失提升视觉生成

AI 摘要

AdvFD 提出了一种新的生成器后训练方法，通过引入可学习的对抗特征空间来补充静态 Fréchet 损失，并采用真实特征白化来稳定优化。实验表明，AdvFD 在 JiT 和 pMF 骨干网络及不同模型规模上均能一致提升一步生成器的后训练效果。该方法旨在解决直接优化 Fréchet 目标时出现的 Fréchet hacking 问题。

正文节选

AdvFD: Boosting Visual Generation via Adversarial Fr'echet Distance Loss Abstract Adversarial Fréchet Distance improves generator post-training by adding a learnable adversarial feature space to static Fréchet losses, with whitening to stabilize optimization. Fréchet distance has recently emerged as an effective distribution-level objective for generator post-training, complementing the conventional sample-level diffusion and flow-matching losses. However, directly optimizing Fréchet objectives

发布时间：—

抓取时间：2026-08-12 13:20

来源机构：Hugging Face