Matryoshka Diffusion Models
First page
Paper summary
Apple introduces an end-to-end framework for high-resolution image and video synthesis that denoises across multiple resolutions jointly.
Ask this paper
01
Joint multi-resolution diffusion: Runs the diffusion process at multiple resolutions simultaneously, sharing representations across scales in a single unified model.
02
NestedUNet: Uses a NestedUNet architecture so that higher-resolution branches build on lower-resolution features without a separate cascade.
03
Progressive training: Trains progressively from low to high resolution, dramatically improving optimization stability for high-resolution generation.
04
Unified model: Eliminates the typical cascaded-diffusion pipeline used in prior high-resolution generation, simplifying training and serving.