ResearchTasksScene Text-Centric Visual Question AnsweringFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedSTVQAVisual-SR10.759Accuracy20May 21, 2026OCRVQAMonkey64.4Accuracy14Feb 27, 2026AI2DDeFacto76.1Accuracy6May 21, 2026