Computer scienceData scienceNLPMain NLP tasksImage recognition as an NLP task

Visual QA

VQA dataset issue

Report a typo

What is the main problem with visual QA datasets?

Select one option from the list

There are very few datasets

Most datasets are designed only to recognize the location

There are no datasets with more than 100k images

Some questions in these datasets are too subjective

___

Create a free account to access the full topic