ResearchDatasetsMS-COCO, CC, VG, SBU, ALTFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsImage CaptioningMS-COCO, CC3M, CC12M, VG, SBU, ALT200M + 0.6B mixed (train)—Primary metric0