There are 0 repository under caption-task topic.
An official implementation for " UniVL: A Unified Video and Language Pre-Training Model for Multimodal Understanding and Generation"