BOOT: Data-free Distillation of Denoising Diffusion Models with Bootstrapping

Understand

Diffusion models have demonstrated excellent potential for generating diverse images.

• However, their performance often suffers from slow generation due to iterative denoising.
• Knowledge distillation has been recently proposed as a remedy that can reduce the number of inference steps to one or a few without significant quality degradation.
• However, existing distillation methods either require significant amounts of offline computation for generating synthetic training data from the teacher model or need to perform expensive online learning with the help of real data.