Improve model card (#2)
Browse files- Improve model card (9164c4f3986626c88af64c652a1bb46923d669a5)
Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -1,8 +1,14 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
|
|
|
| 3 |
---
|
| 4 |
|
| 5 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
|
| 7 |
Website: https://rae-dit.github.io/
|
| 8 |
-
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
pipeline_tag: unconditional-image-generation
|
| 4 |
---
|
| 5 |
|
| 6 |
+
# RAE: Diffusion Transformers with Representation Autoencoders
|
| 7 |
+
|
| 8 |
+
This repository contains the official PyTorch checkpoints for MeanFlow Transformers with Representation Autoencoders (MF-RAE).
|
| 9 |
+
|
| 10 |
+
Representation Autoencoders (RAE) are a class of autoencoders that utilize pretrained, frozen representation encoders such as DINOv2 and SigLIP2 as encoders with trained ViT decoders. RAE can be used in a two-stage training pipeline for high-fidelity image synthesis, where a Stage 2 diffusion model is trained on the latent space of a pretrained RAE to generate images.
|
| 11 |
|
| 12 |
Website: https://rae-dit.github.io/
|
| 13 |
+
Code: https://github.com/bytetriper/RAE
|
| 14 |
+
Paper: https://huggingface.co/papers/2510.11690
|