Update README.md
Browse files
README.md
CHANGED
|
@@ -6,12 +6,13 @@ language:
|
|
| 6 |
- en
|
| 7 |
base_model:
|
| 8 |
- stabilityai/stable-diffusion-xl-base-1.0
|
|
|
|
| 9 |
pipeline_tag: text-to-image
|
| 10 |
---
|
| 11 |
# CHATS: Combining Human-Aligned Optimization and Test-Time Sampling for Text-to-Image Generation (ICML2025)
|
| 12 |
|
| 13 |
<p align="center">
|
| 14 |
-
📝 <a href="https://arxiv.org/abs/2502.12579" target="_blank">Paper</a> • 💡 <a href="https://zhuanlan.zhihu.com/p/1904156278621737628" target="_blank">知乎</a> •
|
| 15 |
</p>
|
| 16 |
|
| 17 |
CHATS is a next-generation framework that unifies human preference alignment with classifier-free guidance by modeling both preferred and dispreferred distributions and using a proxy-prompt-based sampling strategy for superior text–image alignment, fidelity, and aesthetic consistency. See the images generated below for examples.
|
|
|
|
| 6 |
- en
|
| 7 |
base_model:
|
| 8 |
- stabilityai/stable-diffusion-xl-base-1.0
|
| 9 |
+
- madebyollin/sdxl-vae-fp16-fix
|
| 10 |
pipeline_tag: text-to-image
|
| 11 |
---
|
| 12 |
# CHATS: Combining Human-Aligned Optimization and Test-Time Sampling for Text-to-Image Generation (ICML2025)
|
| 13 |
|
| 14 |
<p align="center">
|
| 15 |
+
📝 <a href="https://arxiv.org/abs/2502.12579" target="_blank">Paper</a> • 💡 <a href="https://zhuanlan.zhihu.com/p/1904156278621737628" target="_blank">知乎</a> • 💻 <a href="https://github.com/AIDC-AI/CHATS" target="_blank">Github</a>
|
| 16 |
</p>
|
| 17 |
|
| 18 |
CHATS is a next-generation framework that unifies human preference alignment with classifier-free guidance by modeling both preferred and dispreferred distributions and using a proxy-prompt-based sampling strategy for superior text–image alignment, fidelity, and aesthetic consistency. See the images generated below for examples.
|