Add task category and license to metadata

This PR adds the `task_categories` field to the metadata, specifying `image-segmentation` as the relevant task. It also adds the license information. This improves discoverability and categorization of the dataset on the Hugging Face Hub.

README.md

@@ -37,13 +37,18 @@ dataset_info:
     num_examples: 400
   download_size: 162656888
   dataset_size: 166529577.0
+  license: apache-2.0
 configs:
 - config_name: default
   data_files:
   - split: train
     path: data/train-*
+task_categories:
+- image-segmentation
 ---
 
 **Please see paper & code for more information:**
 - https://github.com/mbzuai-oryx/KITAB-Bench
-- https://arxiv.org/abs/2502.14949
+- https://arxiv.org/abs/2502.14949
+
+KITAB-Bench is a comprehensive multi-domain benchmark for Arabic OCR and document understanding.  It includes tasks such as text recognition (OCR), layout detection, line detection and recognition, table recognition, PDF-to-markdown conversion, chart-to-dataframe conversion, diagram-to-JSON conversion, and visual question answering (VQA).  The dataset contains a wide range of document types from various domains, with high-quality human-labeled annotations.  See the [project website](https://mbzuai-oryx.github.io/KITAB-Bench/) for more details.