search

MILVLG / openvqa

A lightweight, scalable, and general framework for visual question answering research
Apache License 2.0
320 stars 64 forks source link

Question Regarding question-answering architecture #40

Closed qwertyatp closed 5 years ago

qwertyatp commented 5 years ago

Do the models in the model zoo answer only single-word answers, or can they be used for answer generation where the answer consists of multiple words?

Thanks,

MIL-VLG commented 5 years ago

They can be used for generating the answer with multiple words. It is achieved under a classification framework regarding each answer (can be either a single word or multiple words) as an individual class.