This repo contains the official pytorch code for Improving the Diffusability of Autoencoders. Please, refer to the AlphaFlow repo for more details on how to setup the environment and run the training and evaluation pipeline.
TODOs:
- Add the exact instructions for running autoencoder training/evaluation for ImageNet/Kinetics-700.
- Release the checkpoints.
bibtex
@article{skorokhodov2025improving,
title={Improving the diffusability of autoencoders},
author={Skorokhodov, Ivan and Girish, Sharath and Hu, Benran and Menapace, Willi and Li, Yanyu and Abdal, Rameen and Tulyakov, Sergey and Siarohin, Aliaksandr},
journal={arXiv preprint arXiv:2502.14831},
year={2025}
}