Improve model card

This PR improves the model card by:

- Adding a link to the paper
- Changing the pipeline tag to `feature-extraction`
- Adding the license
- Linking to the project page

README.md

@@ -3,7 +3,8 @@ base_model: FacebookAI/xlm-roberta-base
 datasets:
 - StyleDistance/mstyledistance_training_triplets
 library_name: sentence-transformers
-pipeline_tag: sentence-similarity
+pipeline_tag: feature-extraction
+license: mit
 tags:
 - datadreamer
 - datadreamer-0.35.0
@@ -29,9 +30,10 @@ widget:
     меня.
   - 你将使用有限的色调和小尺寸进行像素艺术的简化和风格化设计。
 ---
+
 # Model Card
 
-mStyleDistance is a **multilingual style embedding model** that aims to embed texts with similar writing styles closely and different styles far apart, regardless of content and regardless of language. You may find this model useful for stylistic analysis of multilingual text, clustering, authorship identfication and verification tasks, and automatic style transfer evaluation.
+mStyleDistance is a **multilingual style embedding model** that aims to embed texts with similar writing styles closely and different styles far apart, regardless of content and regardless of language. You may find this model useful for stylistic analysis of multilingual text, clustering, authorship identfication and verification tasks, and automatic style transfer evaluation. The model is described in the paper [StyleDistance/mstyledistance](https://hf.co/papers/2502.15168).
 
 This model is an multilingual version of the English-only [StyleDistance](https://huggingface.co/StyleDistance/styledistance) model.
 
@@ -52,6 +54,9 @@ others = model.encode(["TOCARÁS LA GUITARRA CON TU AMIGO; SERÁ UNA EXCELENTE O
 print(cos_sim(input, others))
 ```
 
+## Project page
+For further information, please check the project page at https://huggingface.co/StyleDistance/mstyledistance.
+
 ---
 ## Trained with DataDreamer