Video-LLaVA【EMNLP 2024】Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionATK3652DEF2429