Add task category and code link

This PR adds the `video-text-to-text` task category to the dataset card, as well as the Github link found at https://github.com/mbzuai-oryx/VideoMolmo.

README.md

@@ -1,6 +1,9 @@
 ---
 license: apache-2.0
+task_categories:
+- video-text-to-text
 ---
+
 # VPoS-Bench: Video Pointing and Segmentation Benchmark
 
 **VPoS-Bench** is a challenging out-of-distribution benchmark designed to evaluate the spatio-temporal pointing and reasoning capabilities of video-language models. It covers a diverse set of five real-world application domains, with fine-grained point-level and segmentation annotations that enable robust evaluation of multimodal models under realistic, temporally complex scenarios.
@@ -11,6 +14,8 @@ license: apache-2.0
 
 > **Model**: [VideoMolmo on Hugging Face](https://huggingface.co/ghazishazan/VideoMolmo)
 
+> **Code**: [VideoMolmo on Github](https://github.com/mbzuai-oryx/VideoMolmo)
+
 ---
 
 ## 🌍 Benchmark Overview
@@ -79,5 +84,4 @@ Each annotation is keyed by a unique video ID and consists of:
       }
     ]
   }
-}
-```
+}