ResearchTasksMulti-modal Task EvaluationMulti-modal Task EvaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMM-MergeBench Seen TasksIndividual83.74SciQA Score12Mar 9, 2026MM-MergeBench Unseen TasksRobustMerge79.3AVQA Score11Mar 9, 2026