ResearchDatasetsSQOOPSQOOPFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsVisual Question AnsweringSQOOP Two objects per image (test)88.5Accuracy4Visual Question AnsweringSQOOP Five objects per image (test)99.96Accuracy4Visual Question AnsweringSQOOP five objects per image (test)99.96Systematic Generalization Accuracy2