view article Article Key Insights into the Law of Vision Representations in MLLMs By Borise • 27 days ago • 16
VISTA: Visualized Text Embedding For Universal Multi-Modal Retrieval Paper • 2406.04292 • Published Jun 6 • 1
MLVU: A Comprehensive Benchmark for Multi-Task Long Video Understanding Paper • 2406.04264 • Published Jun 6 • 1