Многоязычные многомодальные трансформеры
method
ID: multilingualmultimodaltransformers
Related Papers
Delving Deeper into Cross-lingual Visual Question Answering
AbstractVisual question answering (VQA) is one of the crucial vision-and-language tasks. Yet, existing VQA research has mostly focused on the English language, due to a lack of suitable evaluation resources. …...