Logic/DDT
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

update README

Browse Source
This commit is contained in:
wangshuai6
2025-04-09 11:49:03 +08:00
parent 8e37053625
commit b6792002e6
1 changed files with 5 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
README.md
View File

@@ -1,4 +1,9 @@
# DDT: Decoupled Diffusion Transformer
arxiv link: [https://arxiv.org/abs/2504.05741](https://arxiv.org/abs/2504.05741)
ImagNet256 leaderboard: [https://paperswithcode.com/sota/image-generation-on-imagenet-256x256](https://paperswithcode.com/sota/image-generation-on-imagenet-256x256)
ImagNet512 leaderboard: [https://paperswithcode.com/sota/image-generation-on-imagenet-512x512](https://paperswithcode.com/sota/image-generation-on-imagenet-512x512)
## Introduction
We decouple diffusion transformer into encoder-decoder design, and surpresingly that a **more substantial encoder yields performance improvements as model size increases**.
![](./figs/main.png)