CVPR 2024

VCoder: Versatile Vision Encoders for Multimodal Large Language Models