ResearchDatasetsCombined (M3-Bench + Video-MME)FollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsMultimodal Reasoning AverageCombined (M3-Bench + Video-MME-Long)46.8Average Score24