Add model card and metadata

This PR improves the model card for MedShift by:
- Adding the `image-to-image` pipeline tag to the YAML metadata for better discoverability.
- Linking the repository to the original paper, project page, and official GitHub repository.
- Providing a concise summary of the model's purpose: high-fidelity, unpaired cross-domain translation for X-ray images.
- Adding a BibTeX citation for the paper.

README.md

@@ -1,3 +1,26 @@
----
-license: mit
----
+---
+license: mit
+pipeline_tag: image-to-image
+---
+
+# MedShift: Implicit Conditional Transport for X-Ray Domain Adaptation
+
+[**Paper**](https://huggingface.co/papers/2508.21435) | [**Project Page**](https://caetas.github.io/medshift.html) | [**GitHub**](https://github.com/caetas/MedShift)
+
+MedShift is a unified class-conditional generative model based on Flow Matching and Schrödinger Bridges, designed for high-fidelity, unpaired image translation across multiple medical imaging domains. It specifically focuses on bridging the domain gap between synthetic and real X-ray images of the head by addressing discrepancies in attenuation behavior, noise characteristics, and soft tissue representation.
+
+## Description
+Unlike prior approaches that require domain-specific training or rely on paired data, MedShift learns a shared domain-agnostic latent space and supports seamless translation between any pair of domains seen during training. The model is flexible at inference time, as it can be tuned to prioritize either perceptual fidelity or structural consistency, making it a scalable and generalizable solution for domain adaptation in medical imaging.
+
+## Usage
+The official implementation and instructions for installation, training, and inference can be found in the [MedShift GitHub repository](https://github.com/caetas/MedShift).
+
+## Citation
+```bibtex
+@article{caetano2025medshift,
+  title={MedShift: Implicit Conditional Transport for X-Ray Domain Adaptation},
+  author={Caetano, Francisco and Viviers, Christiaan and de With, Peter H. H. and van der Sommen, Fons},
+  journal={arXiv preprint arXiv:2508.21435},
+  year={2025}
+}
+```