LLaVA-Mini: Efficient Image and Video Large Multimodal Models with One Vision Token Paper • 2501.03895 • Published Jan 7 • 53
