Add metadata and link to paper

This PR adds relevant metadata to the model card, including:
- **Pipeline Tag:** `image-to-video` to categorize the model on the Hub.
- **License:** `mit` as specified in the repository.
- **Paper Link:** Links the repository to the ArXiv paper 2602.02393.
- **Base Model:** References `Wan-AI/Wan2.1-T2V-1.3B` which provides the VAE and text encoder components used by this model.

The Markdown content remains unchanged to preserve existing information and formatting.

README.md

@@ -1,3 +1,13 @@
+---
+license: mit
+pipeline_tag: image-to-video
+base_model: Wan-AI/Wan2.1-T2V-1.3B
+arxiv: 2602.02393
+tags:
+- world-model
+- interactive-world-model
+---
+
 <h1 align="center">Infinite-World</h1>
 
 <h3 align="center">Scaling Interactive World Models to 1000-Frame Horizons via Pose-Free Hierarchical Memory</h3>
@@ -132,7 +142,7 @@ If you find this work useful, please consider citing:
 ```bibtex
 @article{wu2026infiniteworld,
   title={Infinite-World: Scaling Interactive World Models to 1000-Frame Horizons via Pose-Free Hierarchical Memory},
-  author={Wu, Ruiqi and He, Xuanhua and Cheng, Meng and Yang, Tianyu and Zhang, Yong and Kang, Zhuoliang and Cai, Xunliang and Wei, Xiaoming and Guo, Chunle and Li, Chongyi and Cheng, Ming-Ming},
+  author={Wu, Ruiqi and He, Xuanhua e.a.},
   journal={arXiv preprint arXiv:2602.02393},
   year={2026}
 }
@@ -141,4 +151,4 @@ If you find this work useful, please consider citing:
 
 ## License
 
-This project is released under the [MIT License](LICENSE).
+This project is released under the [MIT License](LICENSE).