Fetching the paper…

A Joint Study of Phrase Grounding and Task Performance in Vision and Language Models · Around