Fetching the paper…

Multimodal Unified Attention Networks for Vision-and-Language Interactions · Around