Improve model card and add metadata

README.md

@@ -1,3 +1,40 @@
----
-license: cc-by-nc-4.0
----
+---
+license: cc-by-nc-4.0
+pipeline_tag: robotics
+---
+
+# WestWorld: A Knowledge-Encoded Scalable Trajectory World Model for Diverse Robotic Systems
+
+WestWorld is a scalable trajectory world model designed for diverse robotic systems. It addresses scalability challenges using a system-aware Mixture-of-Experts (Sys-MoE) and incorporates domain knowledge of robot physical structures via structural embeddings to enhance zero-shot generalization.
+
+- **Paper:** [WestWorld: A Knowledge-Encoded Scalable Trajectory World Model for Diverse Robotic Systems](https://arxiv.org/abs/2603.14392)
+- **Project Page:** [https://westworldrobot.github.io/](https://westworldrobot.github.io/)
+- **Repository:** [https://github.com/511205787/WestWorld](https://github.com/511205787/WestWorld)
+
+## Evaluation
+
+To evaluate the pretrained model, follow the instructions from the official repository:
+
+1. Place the checkpoint in the `pre_trained/` directory.
+2. Edit `configs/config.yaml` and set the `ckpt_path`. For example:
+```yaml
+ckpt_path: './pre_trained/westworld.ckpt'
+```
+3. Run the evaluation script:
+```bash
+python evaluation_westworld.py
+```
+
+## Citation
+
+If you find WestWorld useful in your research, please cite:
+
+```bibtex
+@inproceedings{wang2026westworld,
+  title={WestWorld: A Knowledge-Encoded Scalable Trajectory World Model for Diverse Robotic Systems},
+  author={Wang, Yuchen and Kong, Jiangtao and Wei, Sizhe and Li, Xiaochang and Lin, Haohong and Zhao, Hongjue and Zhou, Tianyi and Gan, Lu and Shao, Huajie},
+  booktitle={Forty-third International Conference on Machine Learning},
+  year={2026},
+  url={https://openreview.net/forum?id=ncRRCG4BfP}
+}
+```