Add model metadata and links to paper/code

Hi! I'm Niels from the Hugging Face community team. I've noticed this model is missing some important metadata and documentation. This PR adds the `library_name` and `pipeline_tag` to the YAML metadata, which helps with model discoverability and enables automated code snippets. I've also added links to the ICLR 2026 paper and the associated GitHub repository in the README, along with the citation information.

README.md

@@ -1,3 +1,35 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+library_name: transformers
+pipeline_tag: image-text-to-text
+---
+
+# TPRU-7B
+
+This repository contains the model weights for **TPRU-7B**, presented in the paper [TPRU: Advancing Temporal and Procedural Understanding in Large Multimodal Models](https://huggingface.co/papers/2602.18884).
+
+TPRU is a large-scale dataset and training paradigm designed to bridge the gap in Multimodal Large Language Models (MLLMs) regarding temporal and procedural visual data understanding. TPRU-7B leverages reinforcement learning (GRPO) to excel in tasks such as:
+- **Temporal Reordering:** Reconstructing the correct sequence of shuffled frames.
+- **Next-Frame Prediction:** Predicting the immediate future state given a sequence.
+- **Previous-Frame Review:** Deducing the prerequisite state given an outcome.
+
+## Resources
+- **Paper:** [TPRU: Advancing Temporal and Procedural Understanding in Large Multimodal Models](https://huggingface.co/papers/2602.18884)
+- **Code:** [GitHub Repository](https://github.com/Stephen-gzk/TPRU)
+- **Dataset:** [TPRU-25k](https://huggingface.co/datasets/Stephengzk/TPRU-25k)
+
+## Citation
+
+If you find this work useful, please consider citing:
+
+```bibtex
+@inproceedings{gao2026tpru,
+  title={TPRU: Advancing Temporal and Procedural Understanding in Large Multimodal Models},
+  author={Gao, Zhenkun and Wang, Xuhong and Tan, Xin and Xie, Yuan},
+  booktitle={Published as a conference paper at ICLR 2026},
+  year={2026}
+}
+```
+
+## Acknowledgements
+We thank the developers of [Qwen2.5-VL](https://github.com/QwenLM/Qwen2.5-VL), [Easy-R1](https://github.com/hiyouga/EasyR1), and [VLMEvalKit](https://github.com/open-compass/VLMEvalKit) for their open-source contributions.