Add robotics pipeline tag and improve model card

Hi! I'm Niels, part of the community science team at Hugging Face.

This PR improves the model card for the Pi0.5 baseline:
- Adds the `pipeline_tag: robotics` to the metadata to make the model more discoverable on the Hub.
- Organizes links to the paper, project website, and GitHub repositories.
- Adds the BibTeX citation from the paper.

Please review and merge if this looks good to you!

README.md

@@ -1,9 +1,30 @@
 ---
 license: apache-2.0
+pipeline_tag: robotics
 ---
 
 # Finetuned Pi05 Baseline
 
-[Arxiv Paper](https://arxiv.org/abs/2603.04639) |  [HF Paper](https://huggingface.co/papers/2603.04639) |  [Website](https://robomme.github.io/) | [Benchmark Code](https://github.com/RoboMME/robomme_benchmark) | [Policy Learning Code](https://github.com/RoboMME/robomme_policy_learning)
+[Arxiv Paper](https://arxiv.org/abs/2603.04639) | [HF Paper](https://huggingface.co/papers/2603.04639) | [Website](https://robomme.github.io/) | [Benchmark Code](https://github.com/RoboMME/robomme_benchmark) | [Policy Learning Code](https://github.com/RoboMME/robomme_policy_learning)
 
-We fine-tune the pi05 model without history as a baseline on RoboMME. 
+This repository contains the fine-tuned $\pi_{0.5}$ model (without history) used as a baseline for the **RoboMME** benchmark, as presented in the paper [RoboMME: Benchmarking and Understanding Memory for Robotic Generalist Policies](https://huggingface.co/papers/2603.04639).
+
+## About RoboMME
+RoboMME is a large-scale standardized benchmark for evaluating and advancing Vision-Language-Action (VLA) models in long-horizon, history-dependent scenarios. It comprises 16 manipulation tasks evaluating four types of memory:
+- **Temporal memory:** Counting repeated actions.
+- **Spatial memory:** Handling temporarily occluded objects.
+- **Object memory:** Referencing specific objects.
+- **Procedural memory:** Following specific multi-step instructions.
+
+This specific checkpoint serves as a non-memory baseline by using the $\pi_{0.5}$ architecture without historical context.
+
+## Citation
+
+```bibtex
+@article{dai2026robomme,
+  title={RoboMME: Benchmarking and Understanding Memory for Robotic Generalist Policies},
+  author={Dai, Yinpei and Fu, Hongze and Lee, Jayjun and Liu, Yuejiang and Zhang, Haoran and Yang, Jianing and Finn, Chelsea and Fazeli, Nima and Chai, Joyce},
+  journal={arXiv preprint arXiv:2603.04639},
+  year={2026}
+}
+```