Do Vision-and-Language Transformers Learn Grounded Predicate-Noun Dependencies? ... that achieve remarkable performance on multimodal reasoning tasks.
確定! 回上一頁