Add paper link and library name

This PR adds a link to the paper [Task Preference Optimization: Improving Multimodal Large Language Models with Vision Task Alignment](https://huggingface.co/papers/2412.19326).
It also adds `library_name`, enabling people to use the model with the Transformers library.

README.md

@@ -1,12 +1,15 @@
 ---
 base_model:
 - mistralai/Mistral-7B-Instruct-v0.2
+library_name: transformers
 license: mit
 pipeline_tag: video-text-to-text
 ---
 
 # VideoChat2-TPO
 
+This model is based on the paper [Task Preference Optimization: Improving Multimodal Large Language Models with Vision Task Alignment](https://huggingface.co/papers/2412.19326).
+
 ## 🏃 Installation
 
 ```