通过并行梯度蒸馏，实现快速高质扩散模型采样
Distilling Parallel Gradients for Fast ODE Solvers of Diffusion Models
- 每步并行计算多个梯度，降低截断误差
- 5次采样步数下FID低至4.47（CIFAR-10）
- 可作为插件提升现有采样器性能
扩散模型虽生成效果优异，但因逐层去噪导致采样延迟高。现有基于求解器的加速方法常在低延迟预算下牺牲图像质量。本文提出集成并行方向求解器（EPD），通过在每一步中并行评估多个梯度来减少截断误差。由于额外梯度计算相互独立，可完全并行化，保持低延迟。方法以蒸馏方式优化少量可学习参数，训练开销极小。此外，本方法可作为插件提升现有ODE采样器。在多个图像生成基准测试中验证了有效性：在相同5次采样步数（NFE）下，EPD在CIFAR-10上取得FID 4.47，FFHQ为7.97，ImageNet为8.17，LSUN Bedroom为8.26，显著优于现有基于学习的求解器。代码已开源。
原文摘要 · Abstract (English)
Diffusion models (DMs) have achieved state-of-the-art generative performance but suffer from high sampling latency due to their sequential denoising nature. Existing solver-based acceleration methods often face image quality degradation under a low-latency budget. In this paper, we propose the Ensemble Parallel Direction solver (dubbed as \ours), a novel ODE solver that mitigates truncation errors by incorporating multiple parallel gradient evaluations in each ODE step. Importantly, since the additional gradient computations are independent, they can be fully parallelized, preserving low-latency sampling. Our method optimizes a small set of learnable parameters in a distillation fashion, ensuring minimal training overhead. In addition, our method can serve as a plugin to improve existing ODE samplers. Extensive experiments on various image synthesis benchmarks demonstrate the effectiveness of our \ours~in achieving high-quality and low-latency sampling. For example, at the same latency level of 5 NFE, EPD achieves an FID of 4.47 on CIFAR-10, 7.97 on FFHQ, 8.17 on ImageNet, and 8.26 on LSUN Bedroom, surpassing existing learning-based solvers by a significant margin. Codes are available in https://github.com/BeierZhu/EPD.
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。