Improve model card: Add pipeline tag, library name, and Hugging Face paper link

This PR enhances the model card by:
- Promoting the `text-to-image` tag to a top-level `pipeline_tag` for improved discoverability on the Hub.
- Adding `library_name: diffusers` to the metadata, which enables the "How to use" widget with an automated code snippet for `diffusers` usage. This is based on evidence from the repository's `model_index.json` files indicating `diffusers` compatibility.
- Updating the paper links within the model card content to point to the official Hugging Face paper page: [Paris: A Decentralized Trained Open-Weight Diffusion Model](https://huggingface.co/papers/2510.03434).

Please review and merge if these improvements are satisfactory.

README.md

@@ -1,7 +1,8 @@
 ---
 license: mit
+pipeline_tag: text-to-image
+library_name: diffusers
 tags:
-- text-to-image
 - diffusion
 - multi-expert
 - dit
@@ -21,13 +22,13 @@ tags:
 <a href="https://github.com/bageldotcom/paris" target="_blank">
   <img src="https://img.shields.io/badge/⭐_STAR_ON_GITHUB-100000?style=for-the-badge&logo=github&logoColor=white" alt="Star on GitHub" height="40">
 </a>
-<a href="https://github.com/bageldotcom/paris/blob/main/paper.pdf" target="_blank">
+<a href="https://huggingface.co/papers/2510.03434" target="_blank">
   <img src="https://img.shields.io/badge/📄_READ_PAPER-FF6B6B?style=for-the-badge&logoColor=white" alt="Read Technical Report" height="40">
 </a>
 
 <div style="margin-top: 20px;"></div>
 
-The world's first open-weight diffusion model trained entirely through decentralized computation. The model consists of 8 expert diffusion models (129M-605M parameters each) trained in complete isolation with no gradient, parameter, or intermediate activation synchronization, achieving superior parallelism efficiency over traditional methods while using 14× less data and 16× less compute than baselines. [Read our technical report](https://github.com/bageldotcom/paris/blob/main/paper.pdf) to learn more.
+The world's first open-weight diffusion model trained entirely through decentralized computation. The model consists of 8 expert diffusion models (129M-605M parameters each) trained in complete isolation with no gradient, parameter, or intermediate activation synchronization, achieving superior parallelism efficiency over traditional methods while using 14× less data and 16× less compute than baselines. [Read our technical report](https://huggingface.co/papers/2510.03434) to learn more.
 
 # Key Characteristics