DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding Paper • 2412.10302 • Published Dec 13, 2024 • 11
