Fetching the paper…

Learning to Discretely Compose Reasoning Module Networks for Video Captioning · Around