Update README.md

README.md

@@ -51,7 +51,7 @@ This benchmark is from the paper **Narrative-Driven Paper-to-Slide Generation vi
 
 A curated benchmark dataset of **100 oral presentation papers** from top-tier machine learning conferences (CVPR, ICCV, ICLR, ICML, NeurIPS), spanning 2022–2025. Each entry includes the full paper PDF, presentation slides PDF, and rich metadata.
 
-[![arXiv](https://img.shields.io/badge/arXiv-2411.17957-b31b1b.svg)](https://arxiv.org/abs/2411.17957)
+[![arXiv](https://img.shields.io/badge/arXiv-2604.11969-b31b1b.svg)](https://arxiv.org/abs/2604.11969)
 [![Project Page](https://img.shields.io/badge/Project-Page-blue)](https://arcdeck.org/)
 [![Hugging Face](https://img.shields.io/badge/🤗%20Hugging%20Face-Dataset-yellow)](https://huggingface.co/datasets/ArcDeck/ArcBench)
 [![GitHub](https://img.shields.io/badge/GitHub-Repository-black?logo=github)](https://github.com/RehgLab/ArcDeck)