Update readme

README.md

@@ -1,5 +1,5 @@
 ---
-title: Agentic Document VQA Leaderboard
+title: MADQA Leaderboard
 colorFrom: blue
 colorTo: indigo
 sdk: streamlit
@@ -10,9 +10,9 @@ hf_oauth: true
 hf_oauth_expiration_minutes: 43200
 short_description: Navigate, retrieve, and reason over PDFs collection.
 datasets:
-    - agentic-document-ai/dataset
+    - OxRML/MADQA
 ---
 
-# Agentic Document VQA Leaderboard
+# MADQA Leaderboard
 
-This benchmark evaluates AI systems on Agentic Document Collection Visual Question Answering — a task requiring systems to navigate, retrieve, reason over, and aggregate information from heterogeneous document collections.
+Leaderboard for [MADQA (Multimodal Agentic Document QA)](https://huggingface.co/datasets/OxRML/MADQA) — a benchmark of 2,250 human-authored questions over 800 heterogeneous PDF documents, evaluating agentic reasoning over document collections.