Laminating Representation Autoencoders for Efficient Diffusion
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917248312016896 |
|---|---|
| author | Calvo-González, Ramón Fleuret, François |
| author_facet | Calvo-González, Ramón Fleuret, François |
| contents | Recent work has shown that diffusion models can generate high-quality images by operating directly on SSL patch features rather than pixel-space latents. However, the dense patch grids from encoders like DINOv2 contain significant redundancy, making diffusion needlessly expensive. We introduce FlatDINO, a variational autoencoder that compresses this representation into a one-dimensional sequence of just 32 continuous tokens -an 8x reduction in sequence length and 48x compression in total dimensionality. On ImageNet 256x256, a DiT-XL trained on FlatDINO latents achieves a gFID of 1.80 with classifier-free guidance while requiring 8x fewer FLOPs per forward pass and up to 4.5x fewer FLOPs per training step compared to diffusion on uncompressed DINOv2 features. These are preliminary results and this work is in progress. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_04873 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Laminating Representation Autoencoders for Efficient Diffusion Calvo-González, Ramón Fleuret, François Computer Vision and Pattern Recognition Recent work has shown that diffusion models can generate high-quality images by operating directly on SSL patch features rather than pixel-space latents. However, the dense patch grids from encoders like DINOv2 contain significant redundancy, making diffusion needlessly expensive. We introduce FlatDINO, a variational autoencoder that compresses this representation into a one-dimensional sequence of just 32 continuous tokens -an 8x reduction in sequence length and 48x compression in total dimensionality. On ImageNet 256x256, a DiT-XL trained on FlatDINO latents achieves a gFID of 1.80 with classifier-free guidance while requiring 8x fewer FLOPs per forward pass and up to 4.5x fewer FLOPs per training step compared to diffusion on uncompressed DINOv2 features. These are preliminary results and this work is in progress. |
| title | Laminating Representation Autoencoders for Efficient Diffusion |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.04873 |