Update README.md

README.md

@@ -1,3 +1,93 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+---
+language:
+- en
+pipeline_tag: text-to-image
+---
+# BizGen: Advancing Article-level Visual Text Rendering for Infographics Generation
+
+<a href="https://arxiv.org/abs/2406.04314"><img src="https://img.shields.io/badge/Paper-arXiv-red?style=for-the-badge" height=22.5></a>
+<a href="https://github.com/RockeyCoss/SPO"><img src="https://img.shields.io/badge/Gihub-Code-succees?style=for-the-badge&logo=GitHub" height=22.5></a>
+<a href="https://rockeycoss.github.io/spo.github.io/"><img src="https://img.shields.io/badge/Project-Page-blue?style=for-the-badge" height=22.5></a>
+
+<table>
+  <tr>
+    <td><img src="assets/imgs/0.png" alt="teaser example 0" width="200"/></td>
+    <td><img src="assets/imgs/1.png" alt="teaser example 1" width="200"/></td>
+    <td><img src="assets/imgs/2.png" alt="teaser example 2" width="200"/></td>
+    <td><img src="assets/imgs/3.png" alt="teaser example 3" width="200"/></td>
+  </tr>
+</table>
+
+## Abstract
+<p>
+    Generating visually appealing images is fundamental to modern text-to-image generation models. 
+    A potential solution to better aesthetics is direct preference optimization (DPO), 
+    which has been applied to diffusion models to improve general image quality including prompt alignment and aesthetics. 
+    Popular DPO methods propagate preference labels from clean image pairs to all the intermediate steps along the two generation trajectories. 
+    However, preference labels provided in existing datasets are blended with layout and aesthetic opinions, which would disagree with aesthetic preference. 
+    Even if aesthetic labels were provided (at substantial cost), it would be hard for the two-trajectory methods to capture nuanced visual differences at different steps.
+</p>
+<p>
+    To improve aesthetics economically, this paper uses existing generic preference data and introduces step-by-step preference optimization 
+    (SPO) that discards the propagation strategy and allows fine-grained image details to be assessed. Specifically, 
+    at each denoising step, we 1) sample a pool of candidates by denoising from a shared noise latent, 
+    2) use a step-aware preference model to find a suitable win-lose pair to supervise the diffusion model, and 
+    3) randomly select one from the pool to initialize the next denoising step. 
+    This strategy ensures that diffusion models focus on the subtle, fine-grained visual differences 
+    instead of layout aspect. We find that aesthetic can be significantly enhanced by accumulating these 
+    improved minor differences.
+</p>
+<p>
+    When fine-tuning Stable Diffusion v1.5 and SDXL, SPO yields significant 
+    improvements in aesthetics compared with existing DPO methods while not sacrificing image-text alignment 
+    compared with vanilla models. Moreover, SPO converges much faster than DPO methods due to the step-by-step 
+    alignment of fine-grained visual details.
+</p>
+
+## Model Description
+
+This model is fine-tuned from [stable-diffusion-xl-base-1.0](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0). It has been trained on 4,000 prompts for 10 epochs.
+
+This is a merged checkpoint that combines the LoRA checkpoint with the base model [stable-diffusion-xl-base-1.0](https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0). If you want to access the LoRA checkpoint, please visit [SPO-SDXL_4k-p_10ep_LoRA](https://huggingface.co/SPO-Diffusion-Models/SPO-SDXL_4k-p_10ep_LoRA). We also provide a LoRA checkpoint compatible with [stable-diffusion-webui](https://github.com/AUTOMATIC1111/stable-diffusion-webui), which can be accessed [here](https://civitai.com/models/510261?modelVersionId=567119).
+
+
+## A quick example
+```python
+from diffusers import StableDiffusionXLPipeline, UNet2DConditionModel
+import torch
+
+# load pipeline
+inference_dtype = torch.float16
+pipe = StableDiffusionXLPipeline.from_pretrained(
+    "SPO-Diffusion-Models/SPO-SDXL_4k-p_10ep", 
+    torch_dtype=inference_dtype,
+)
+vae = AutoencoderKL.from_pretrained(
+    'madebyollin/sdxl-vae-fp16-fix',
+    torch_dtype=inference_dtype,
+)
+pipe.vae = vae
+pipe.to('cuda')
+
+generator=torch.Generator(device='cuda').manual_seed(42)
+image = pipe(
+    prompt='a child and a penguin sitting in front of the moon',
+    guidance_scale=5.0,
+    generator=generator,
+    output_type='pil',
+).images[0]
+image.save('moon.png')
+```
+
+## Citation
+If you find our work or codebase useful, please consider giving us a star and citing our work.
+```
+@article{liang2024step,
+  title={Aesthetic Post-Training Diffusion Models from Generic Preferences with Step-by-step Preference Optimization},
+  author={Liang, Zhanhao and Yuan, Yuhui and Gu, Shuyang and Chen, Bohan and Hang, Tiankai and Cheng, Mingxi and Li, Ji and Zheng, Liang},
+  journal={arXiv preprint arXiv:2406.04314},
+  year={2024}
+}
+```