Add task category and improve dataset card

Hi, I'm Niels from the community science team at Hugging Face. This PR improves the dataset card for MapTab by:
- Adding the `image-text-to-text` task category to the metadata for better discoverability.
- Adding links to the paper, project page, and official GitHub repository.
- Organizing the description to clearly outline the Metromap and Travelmap scenarios.
- Adding sample usage scripts and instructions as found in the GitHub README.
- Including the BibTeX citation.

README.md

@@ -1,23 +1,64 @@
+---
+task_categories:
+- image-text-to-text
+---
+
 # MapTab
 
-This repository contains the dataset described in the paper **MapTab: Are MLLMs Ready for Multi-Criteria Route Planning in Heterogeneous Graphs?**
+MapTab is a multimodal benchmark specifically designed to evaluate holistic multi-criteria reasoning in Multimodal Large Language Models (MLLMs) via route planning tasks. It requires MLLMs to perceive and ground visual cues from map images alongside route attributes (e.g., Time, Price) from structured tabular data.
 
-- Paper: https://arxiv.org/abs/2602.18600
-- Project page: https://ziqiao-shang.github.io/MapTab-Leaderboard/
-- Code: https://github.com/Ziqiao-Shang/MapTab
+- **Paper:** [MapTab: Are MLLMs Ready for Multi-Criteria Route Planning in Heterogeneous Graphs?](https://arxiv.org/abs/2602.18600)
+- **Project Page:** [https://ziqiao-shang.github.io/MapTab-Leaderboard/](https://ziqiao-shang.github.io/MapTab-Leaderboard/)
+- **Code:** [https://github.com/Ziqiao-Shang/MapTab](https://github.com/Ziqiao-Shang/MapTab)
 
-## Usage
+## Dataset Description
 
-Please copy the released **`tabulars`**, **`images`**, and **`queries`** folders into the corresponding folders in the code repository:
+The benchmark encompasses two scenarios:
+- **Metromap**: Covers metro networks in 160 cities across 52 countries.
+- **Travelmap**: Depicts 168 representative tourist attractions from 19 countries.
 
-- `metromap/tabulars`, `metromap/images`, `metromap/qa`
-- `travelmap/tabulars`, `travelmap/images`, `travelmap/qa`
+In total, MapTab comprises 328 images, 196,800 route planning queries, and 3,936 QA queries, all incorporating 4 key criteria: Time, Price, Comfort, and Reliability.
 
 ## Current Release
 
 Currently, we only release the **`test_set` for the route planning task** for evaluation.
 
 The following parts will be released in future updates:
-
 - route planning **`training_set`**
-- QA task **test_sets**
+- QA task **test_sets**
+
+## Usage
+
+### Setup
+Please copy the released **`tabulars`**, **`images`**, and **`queries`** folders into the corresponding folders in the [official code repository](https://github.com/Ziqiao-Shang/MapTab):
+
+- `metromap/tabulars`, `metromap/images`, `metromap/qa`
+- `travelmap/tabulars`, `travelmap/images`, `travelmap/qa`
+
+### Quick Start
+You can run generation and evaluation using the scripts provided in the repository:
+
+```bash
+# Generate RP task results
+bash scripts/generate_rp.sh
+
+# Generate QA task results
+bash scripts/generate_qa.sh
+
+# Evaluate RP task results
+bash scripts/evaluate_rp.sh
+
+# Evaluate QA task results
+bash scripts/evaluate_qa.sh
+```
+
+## Citation
+
+```bibtex
+@article{shang2026maptab,
+  title={MapTab: Can MLLMs Master Constrained Route Planning?},
+  author={Shang, Ziqiao and Ge, Lingyue and Chen, Yang and Tian, Shi-Yu and Huang, Zhenyu and Fu, Wenbo and Li, Yu-Feng and Guo, Lan-Zhe},
+  journal={arXiv preprint arXiv:2602.18600},
+  year={2026}
+}
+```