Accelerating Diffusion Sampling with Classifier-based Feature Distillation
This work addresses a practical bottleneck for users of diffusion models in image generation, though it is incremental over existing distillation methods.
The paper tackles the slow sampling speed of diffusion models by proposing Classifier-based Feature Distillation (CFD), which improves few-step samplers by distilling teacher features with a classifier, achieving high-quality and fast sampling on CIFAR-10.
Although diffusion model has shown great potential for generating higher quality images than GANs, slow sampling speed hinders its wide application in practice. Progressive distillation is thus proposed for fast sampling by progressively aligning output images of $N$-step teacher sampler with $N/2$-step student sampler. In this paper, we argue that this distillation-based accelerating method can be further improved, especially for few-step samplers, with our proposed \textbf{C}lassifier-based \textbf{F}eature \textbf{D}istillation (CFD). Instead of aligning output images, we distill teacher's sharpened feature distribution into the student with a dataset-independent classifier, making the student focus on those important features to improve performance. We also introduce a dataset-oriented loss to further optimize the model. Experiments on CIFAR-10 show the superiority of our method in achieving high quality and fast sampling. Code is provided at \url{https://github.com/zju-SWJ/RCFD}.