Update README.md

README.md

@@ -7,24 +7,30 @@
   <a href="https://github.com/SkyworkAI/Matrix-Game">
     <img src="https://img.shields.io/badge/GitHub-100000?style=flat&logo=github&logoColor=white" alt="GitHub">
   </a>
-  <a href="#todo">
+  <a href="https://github.com/SkyworkAI/Matrix-Game/blob/main/assets/report.pdf">
     <img src="https://img.shields.io/badge/arXiv-Report-b31b1b?style=flat&logo=arxiv&logoColor=white" alt="arXiv">
   </a>
+
+  
 </div>
 
 ## 📝 Overview
-**Matrix-Game** is a 17B-parameter Diffusion Transformer for generating high-resolution, physics-consistent videos in interactive game environments. Trained on large-scale data from Minecraft and Unreal Engine, it understands game physics like collisions, destruction, and item placement. Matrix-Game supports real-time, action-conditioned generation, adapting video content dynamically to user input.
+**Matrix-Game** is a 17B-parameter interactive world foundation model for controllable game world generation.
+
+## ✨ Key Features
 
-You can find more visualizations on our [website](#).
+- 🎯 **Feature 1**: **Interactive Generation.**  A diffusion-based image-to-world model that generates high-quality videos conditioned on keyboard and mouse inputs, enabling fine-grained control and dynamic scene evolution.
+- 🚀 **Feature 2**: **GameWorld Score.** A comprehensive benchmark for evaluating Minecraft world models across four key dimensions, including visual quality, temporal quality, action controllability, and physical rule understanding. 
+- 💡 **Feature 3**: **Matrix-Game Dataset** A large-scale Minecraft dataset with fine-grained action annotations, supporting scalable training for interactive and physically grounded world modeling.
 
 ## 🔥 Latest Updates
 
-* [2025-05] 🎉 Initial release of Matrix-Game
+* [2025-05] 🎉 Initial release of Matrix-Game Model
 
 ## 🚀 Performance Comparison
 ### GameWorld Score Benchmark Comparison
 
-| Model     | Image Quality ↑ | Aesthetic ↑ | Temporal Cons. ↑ | Motion Smooth. ↑ | Keyboard Acc. ↑ | Mouse Acc. ↑ | 3D Cons. ↑ |
+| Model     | Image Quality ↑ | Aesthetic Quality ↑ | Temporal Cons. ↑ | Motion Smooth. ↑ | Keyboard Acc. ↑ | Mouse Acc. ↑ | 3D Cons. ↑ |
 |-----------|------------------|-------------|-------------------|-------------------|------------------|---------------|-------------|
 | Oasis     | 0.65             | 0.48        | 0.94              | **0.98**          | 0.77             | 0.56          | 0.56        |
 | MineWorld | 0.69             | 0.47        | 0.95              | **0.98**          | 0.86             | 0.64          | 0.51        |
@@ -33,128 +39,57 @@ You can find more visualizations on our [website](#).
 **Metric Descriptions**:
 
 - **Image Quality** / **Aesthetic**: Visual fidelity and perceptual appeal of generated frames  
-- **Temporal Cons.** / **Motion Smooth.**: Temporal coherence and smoothness between frames  
-- **Keyboard Acc.** / **Mouse Acc.**: Accuracy in following user control signals  
-- **3D Cons.**: Geometric stability and physical plausibility over time
+- **Temporal Consistency** / **Motion Smoothness**: Temporal coherence and smoothness between frames  
+- **Keyboard Accuracy** / **Mouse Accuracy**: Accuracy in following user control signals  
+- **3D Consistency**: Geometric stability and physical plausibility over time
+
+  Please check our [GameWorld](https://github.com/SkyworkAI/Matrix-Game/tree/main/GameWorldScore) benchmark for detailed implementation.
 
 ### Human Evaluation
-<table>
-  <thead>
-    <tr>
-      <th>Group</th>
-      <th>Method</th>
-      <th>Overall Quality (%)</th>
-      <th>Controllability (%)</th>
-      <th>Visual Quality (%)</th>
-      <th>Temporal Consistency (%)</th>
-    </tr>
-  </thead>
-  <tbody>
-    <tr>
-      <td rowspan="3">Group A</td>
-      <td>Oasis</td>
-      <td>0.16</td>
-      <td>0.33</td>
-      <td>0.00</td>
-      <td>0.16</td>
-    </tr>
-    <tr>
-      <td>MineWorld</td>
-      <td>3.78</td>
-      <td>5.58</td>
-      <td>1.32</td>
-      <td>13.82</td>
-    </tr>
-    <tr>
-      <td><strong>Ours</strong></td>
-      <td><strong>96.05</strong></td>
-      <td><strong>94.09</strong></td>
-      <td><strong>98.68</strong></td>
-      <td><strong>86.02</strong></td>
-    </tr>
-    <tr>
-      <td rowspan="3">Group B</td>
-      <td>Oasis</td>
-      <td>0.66</td>
-      <td>0.82</td>
-      <td>0.75</td>
-      <td>0.66</td>
-    </tr>
-    <tr>
-      <td>MineWorld</td>
-      <td>2.79</td>
-      <td>5.76</td>
-      <td>1.48</td>
-      <td>6.25</td>
-    </tr>
-    <tr>
-      <td><strong>Ours</strong></td>
-      <td><strong>96.55</strong></td>
-      <td><strong>93.42</strong></td>
-      <td><strong>97.77</strong></td>
-      <td><strong>93.09</strong></td>
-    </tr>
-    <tr>
-      <td rowspan="3">Average</td>
-      <td>Oasis</td>
-      <td>0.41</td>
-      <td>0.58</td>
-      <td>0.38</td>
-      <td>0.41</td>
-    </tr>
-    <tr>
-      <td>MineWorld</td>
-      <td>3.29</td>
-      <td>5.67</td>
-      <td>1.40</td>
-      <td>10.04</td>
-    </tr>
-    <tr>
-      <td><strong>Ours</strong></td>
-      <td><strong>96.30</strong></td>
-      <td><strong>93.76</strong></td>
-      <td><strong>98.23</strong></td>
-      <td><strong>89.56</strong></td>
-    </tr>
-  </tbody>
-</table>
+
+![Human Win Rate](https://raw.githubusercontent.com/SkyworkAI/Matrix-Game/main/assets/imgs/human_win_rate.png)
 
 > Double-blind human evaluation by two independent groups across four key dimensions: **Overall Quality**, **Controllability**, **Visual Quality**, and **Temporal Consistency**.  
 > Scores represent the percentage of pairwise comparisons in which each method was preferred. Matrix-Game consistently outperforms prior models across all metrics and both groups.
 
 
-## 🛠️ Installation
+## 🚀 Quick Start
 
-1. Clone the repository:
-```bash
+```
+# clone the repository:
 git clone https://github.com/SkyworkAI/Matrix-Game.git
 cd Matrix-Game
-```
 
-2. Install dependencies:
-```bash
+# install dependencies:
 pip install -r requirements.txt
-```
 
-## 🚀 Quick Start
+# install apex and FlashAttention-3
+# Our project also depends on [apex](https://github.com/NVIDIA/apex) and [FlashAttention-3](https://github.com/Dao-AILab/flash-attention)
 
-```bash
+# inference
 bash run_inference.sh
 ```
 
-## 🤝 Contributing
-
-We welcome contributions! Please see our [contributing guidelines](CONTRIBUTING.md) for more details.
-
 ## ⭐ Acknowledgements
 
 We would like to express our gratitude to:
 
 - [Diffusers](https://github.com/huggingface/diffusers) for their excellent diffusion model framework
 - [HunyuanVideo](https://github.com/Tencent/HunyuanVideo) for their strong base model
+- [MineDojo](https://minedojo.org/knowledge_base) for their Minecraft video dataset
+- [MineRL](https://github.com/minerllabs/minerl) for their excellent gym framework
+- [Video-Pre-Training](https://github.com/openai/Video-Pre-Training) for their accurate Inverse Dynamics Model
+- [GameFactory](https://github.com/KwaiVGI/GameFactory) for their idea of action control module 
 
 We are grateful to the broader research community for their open exploration and contributions to the field of interactive world generation.
 
-## 📄 License
-
-This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
+## 📎 Citation
+If you find this project useful, please cite our paper:
+```bibtex
+@article{zhang2025matrixgame,
+  title     = {Matrix-Game: Interactive World Foundation Model},
+  author    = {Yifan Zhang and Chunli Peng and Boyang Wang and Puyi Wang and Qingcheng Zhu and Zedong Gao and Eric Li and Yang Liu and Yahui Zhou},
+  journal   = {arXiv},
+  year      = {2025}
+}
+```