UniT: Multimodal Multitask Learning with a Unified Transformer.

Ronghang Hu, Amanpreet Singh

Browse the full ICCV paper archive.