Improve model card: Correct pipeline tag, add library name, and link project page

This PR improves the model card by:

- Correcting the `pipeline_tag` to `video-text-to-text`.
- Adding `library_name: transformers`
- Adding a link to the project page.

This update enhances discoverability and provides a more comprehensive description of the model.

README.md

@@ -1,6 +1,13 @@
 ---
-license: mit
 base_model:
 - Qwen/Qwen2.5-VL-3B-Instruct
-pipeline_tag: visual-question-answering
----
+license: mit
+pipeline_tag: video-text-to-text
+library_name: transformers
+---
+
+This repository contains the model described in [Spatial-MLLM: Boosting MLLM Capabilities in Visual-based Spatial Intelligence](https://huggingface.co/papers/2505.23747).
+
+Project page: https://diankun-wu.github.io/Spatial-MLLM/
+
+Code: https://github.com/diankun-wu/Spatial-MLLM