给扩散模型加对抗训练，让生成过程更稳更抗干扰。
What is Adversarial Training for Diffusion Models?
- 用对抗噪声保持扩散过程与数据分布一致，而非单纯不变输出
- 在低维和高维数据上均显著提升对噪声、异常值的鲁棒性
- 无需预设噪声模型，可无缝集成到现有扩散训练流程
本文回答了扩散模型对抗训练的本质问题：与分类器中要求输出不变不同，扩散模型的对抗训练需保持过程等变性，以确保扩散流与数据分布对齐。该方法通过引入随机噪声或对抗噪声，实现扩散过程的平滑，增强对异常值和污染数据的鲁棒性。无需假设噪声模型，可自然融入扩散训练，具备处理噪声数据、应对极端变化、防止记忆过拟合及提升整体鲁棒性的内在能力。我们在具有已知分布的低维与高维合成数据集上进行严格验证，准确测量误差；并在CIFAR-10、CelebA和LSUN Bedroom标准基准上测试，展示在严重噪声、数据污染和迭代对抗攻击下仍保持优异性能。
原文摘要 · Abstract (English)
We answer the question in the title, showing that adversarial training (AT) for diffusion models (DMs) fundamentally differs from classifiers: while AT in classifiers enforces output invariance, AT in DMs requires equivariance to keep the diffusion process aligned with the data distribution. AT is a way to enforce smoothness in the diffusion flow, improving robustness to outliers and corrupted data. Unlike prior art, our method makes no assumptions about the noise model and integrates seamlessly into diffusion training by adding random noise, similar to randomized smoothing, or adversarial noise, akin to AT. This enables intrinsic capabilities such as handling noisy data, dealing with extreme variability such as outliers, preventing memorization, and improving robustness. We rigorously evaluate our approach with proof-of-concept datasets with known distributions in low- and high-dimensional space, thereby taking a perfect measure of errors; we further evaluate on standard benchmarks such as CIFAR-10, CelebA and LSUN Bedroom, showing strong performance under severe noise, data corruption, and iterative adversarial attacks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。