Improve model card and add robotics tag

README.md

@@ -1,7 +1,16 @@
+---
+pipeline_tag: robotics
+---
+
 # Light-WAM Checkpoints
 
 This repository contains released checkpoints for [**Light-WAM: Efficient World Action Models with State-Fusion Action Decoding**](https://arxiv.org/abs/2606.08242).
 
+Light-WAM is a lightweight World Action Model for efficient robot manipulation. It leverages a compact video backbone (Wan2.1-T2V-1.3B) and performs future-video supervision in a downsampled latent space, allowing it to maintain strong performance on LIBERO and RoboTwin 2.0 while using only 0.44B trainable parameters.
+
+- **Code:** [GitHub Repository](https://github.com/L1ziang/Light-WAM)
+- **Paper:** [arXiv:2606.08242](https://arxiv.org/abs/2606.08242)
+
 ## Contents
 
 Released checkpoints:
@@ -36,4 +45,20 @@ Download all Light-WAM checkpoints:
 ```bash
 hf download l1ziang/lightwam-checkpoints \
   --local-dir lightwam-checkpoints
+```
+
+## Citation
+
+If you find Light-WAM useful in your research, please cite:
+
+```bibtex
+@misc{li2026lightwam,
+  title         = {Light-WAM: Efficient World Action Models with State-Fusion Action Decoding},
+  author        = {Ziang Li and Dongzhou Cheng and Yibin Wang and Shiyue Wang and Xiaoyang Xu and Lingxuan Weng and Juan Wang and Jiaqi Wang},
+  year          = {2026},
+  eprint        = {2606.08242},
+  archivePrefix = {arXiv},
+  primaryClass  = {cs.RO},
+  url           = {https://arxiv.org/abs/2606.08242}
+}
 ```