ResearchTasksImage-Language UnderstandingImage-Language UnderstandingFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedSQAOriginal71.6EM21Feb 26, 2026SEEDBOriginal69.9Accuracy21Feb 26, 2026MMBenchVanilla67.01Score8Feb 26, 2026COCOVanilla1Score8Feb 26, 2026