Add dataset card, link to paper and project page

Hi! I'm Niels from the Hugging Face community science team. I'm opening this PR to improve the dataset card for the OmniViTac dataset. This update includes:
- Adding the `robotics` task category to the metadata.
- Linking the dataset to the original research paper, project page, and GitHub repository.
- Providing a brief description of the dataset's scope and content based on the paper abstract.

This information helps researchers better understand and cite the dataset.

README.md

@@ -1,3 +1,38 @@
----
-license: cc-by-nc-4.0
----
+---
+license: cc-by-nc-4.0
+task_categories:
+- robotics
+tags:
+- visuo-tactile
+- manipulation
+- world-models
+---
+
+# OmniViTac Dataset
+
+OmniViTac is a large-scale visuo-tactile-action dataset introduced in the paper [OmniVTA: Visuo-Tactile World Modeling for Contact-Rich Robotic Manipulation](https://huggingface.co/papers/2603.19201).
+
+[**Project Page**](https://mrsecant.github.io/OmniVTA) | [**GitHub**](https://github.com/MrSecant/OmniVTA) | [**Paper**](https://huggingface.co/papers/2603.19201)
+
+## Dataset Summary
+The OmniViTac dataset comprises 21,000+ trajectories across 86 tasks and 100+ objects, organized into six physics-grounded interaction patterns. It is designed to support the development of world-model-based visuo-tactile manipulation frameworks, providing high-frequency tactile feedback alongside visual observations and robot actions for contact-rich tasks such as wiping and assembly.
+
+## Interaction Patterns
+The dataset covers six physics-grounded interaction categories:
+1. Wiping
+2. Assembly
+3. Pushing
+4. Sliding
+5. Grasping
+6. Peg-in-hole insertion
+
+## Citation
+If you use this dataset in your research, please cite the following paper:
+```bibtex
+@article{OmniVTA2026,
+  title={OmniVTA: Visuo-Tactile World Modeling for Contact-Rich Robotic Manipulation},
+  author={Anonymous},
+  journal={arXiv preprint arXiv:2603.19201},
+  year={2026}
+}
+```