Update dataset card with paper, code, and project links

README.md

@@ -1,10 +1,11 @@
 ---
 license: cc-by-4.0
-pretty_name: TexasPokerRobot
 size_categories:
 - 1K<n<10K
-tags:
+pretty_name: TexasPokerRobot
+task_categories:
 - robotics
+tags:
 - robot-learning
 - imitation-learning
 - manipulation
@@ -16,17 +17,21 @@ configs:
     path: data/train.csv
 ---
 
-# TexasPokerRobot
+# TexasPokerRobot (DexHoldem)
 
-TexasPokerRobot is a robot manipulation dataset collected in a Texas poker tabletop environment. The raw episodes are stored as compressed NumPy `.npz` files, organized by action folder. This release adds a Hugging Face-compatible manifest at `data/train.csv` so the dataset has a standard loadable split and a working Dataset Viewer while preserving the original raw episode files.
+[**Project Page**](https://dexholdem.github.io/Dexholdem/) | [**Paper**](https://huggingface.co/papers/2605.18727) | [**Code**](https://github.com/DexHoldem/Dexholdem-Policy)
+
+TexasPokerRobot is a robot manipulation dataset collected in a Texas poker tabletop environment, introduced in the paper "DexHoldem: Playing Texas Hold'em with Dexterous Embodied System". The dataset provides 1,470 teleoperated demonstrations across 14 Texas Hold'em manipulation primitives using a ShadowHand and UR arm. 
+
+The raw episodes are stored as compressed NumPy `.npz` files, organized by action folder. This release adds a Hugging Face-compatible manifest at `data/train.csv` so the dataset has a standard loadable split and a working Dataset Viewer while preserving the original raw episode files.
 
 ## Dataset Summary
 
-- 1,470 raw episode files
-- 14 action folders, with 105 episodes per action
-- 377.94 GB of raw `.npz` episode data
-- RGB observations from three cameras, depth observations from three cameras, and robot joint state streams
-- License: CC BY 4.0
+- **Total Episodes**: 1,470 raw episode files
+- **Action Diversity**: 14 action folders, with 105 episodes per action
+- **Total Size**: 377.94 GB of raw `.npz` episode data
+- **Observations**: RGB observations from three cameras, depth observations from three cameras, and robot joint state streams
+- **License**: CC BY 4.0
 
 ## Actions
 
@@ -102,3 +107,14 @@ This dataset is intended for robot learning research, imitation learning, action
 ## Limitations
 
 The Dataset Viewer previews the manifest rows, not the full RGB, depth, and joint-state tensors. The raw episode files are large, and no official train/validation/test benchmark split is provided beyond the manifest split.
+
+## Citation
+
+```bibtex
+@article{chen2026dexholdem,
+  title={DexHoldem: Playing Texas Hold'em with Dexterous Embodied System},
+  author={Feng Chen and Tianzhe Chu and Li Sun and Pei Zhou and Zhuxiu Xu and Shenghua Gao and Yuexiang Zhai and Yanchao Yang and Yi Ma},
+  journal={arXiv preprint arXiv:2605.18727},
+  year={2026}
+}
+```