Update model card metadata, links and usage instructions

README.md

@@ -1,14 +1,18 @@
 ---
-license: mit
+language:
+- en
 library_name: diffusers
+license: mit
 pipeline_tag: text-to-image
+base_model: black-forest-labs/FLUX.1-dev
+datasets:
+- FudanCVL/SA-Z
 tags:
 - occlusion
 - layout
 - flux
-language:
-- en
 ---
+
 <table align="center" cellspacing="0" cellpadding="0" style="margin:0 auto;">
   <tr>
     <td valign="middle" style="padding-right:10px;">
@@ -23,9 +27,9 @@ language:
 
 <div align="center" style="margin-top:14px;">
 <a href='https://henghuiding.com/OcclusionFormer/'><img src='https://img.shields.io/badge/Project-Page-Green'></a> &nbsp;&nbsp;&nbsp;&nbsp;
-<a href='https://icml.cc/Downloads/2026'><img src='https://img.shields.io/badge/ICML-2026-blue'></a> &nbsp;&nbsp;&nbsp;&nbsp;
-<a href='https://arxiv.org/'><img src='https://img.shields.io/badge/arXiv-Coming%20Soon-b31b1b'></a> &nbsp;&nbsp;&nbsp;&nbsp;
-<a href='https://huggingface.co/'><img src='https://img.shields.io/badge/%F0%9F%A4%97%20Weights-Hugging Face%20-orange'></a> &nbsp;&nbsp;&nbsp;&nbsp;
+<a href='https://icml.cc/virtual/2026/poster/66159'><img src='https://img.shields.io/badge/ICML-2026-blue'></a> &nbsp;&nbsp;&nbsp;&nbsp;
+<a href='https://arxiv.org/abs/2605.21343'><img src='https://img.shields.io/badge/arXiv-2605.21343-b31b1b.svg'></a> &nbsp;&nbsp;&nbsp;&nbsp;
+<a href='https://huggingface.co/papers/2605.21343'><img src='https://img.shields.io/badge/%F0%9F%A4%97%20Paper-Hugging Face-orange'></a> &nbsp;&nbsp;&nbsp;&nbsp;
 <a href='https://huggingface.co/datasets/FudanCVL/SA-Z'><img src='https://img.shields.io/badge/%F0%9F%A4%97%20SA--Z-Hugging Face%20-orange'></a>
 </div>
 <p align="center" style="margin:4px 0 0 0;">
@@ -36,7 +40,12 @@ language:
 <p align="center" style="margin:1px 0 0 0; font-size:1.48em; font-weight:900; color:#ff6a00;">ICML 2026</p>
 <p align="center" style="margin:1px 0 0 0; font-size:1.08em; color:#6b7280;"><em>✉ Corresponding Author</em></p>
 
+This repository contains the weights for **OcclusionFormer**, as presented in the paper [OcclusionFormer: Arranging Z-Order for Layout-Grounded Image Generation](https://huggingface.co/papers/2605.21343).
+
+The official code is available at: [https://github.com/FudanCVL/OcclusionFormer](https://github.com/FudanCVL/OcclusionFormer)
+
 ## 🔥 News
+- [2026/05/21] Release **OcclusionFormer** on [ArXiv](https://arxiv.org/abs/2605.21343)!
 - [2026/05/18] Release **inference code**, **model weights** and **SA-Z dataset**.
 - [2026/05/18] Release **OcclusionFormer open-source package** in this repository.
 - [2026/4/30] OcclusionFormer is accepted to **ICML 2026**.
@@ -85,9 +94,7 @@ pip install --upgrade -r requirements.txt
 
 3. Download checkpoint
 
-```bash
-https://huggingface.co/FudanCVL/OcclusionFormer
-```
+Weights can be found in this repository. Download `occlusionformer` to `./ckpt`.
 
 4. Run Streamlit demo
 
@@ -107,18 +114,6 @@ python inference_occlusionformer.py \
   --overwrite
 ```
 
-Batch inference with a directory of JSON layouts:
-
-```bash
-python inference_occlusionformer.py \
-  --model_path /path/to/FLUX.1-dev \
-  --ckpt_path /path/to/occlusionformer_checkpoint_dir \
-  --layout_dir ./examples \
-  --output_dir ./outputs_occlusionformer \
-  --enable_layout \
-  --overwrite
-```
-
 ---
 ## ✅ TODO