ResearchDatasetsVideoDetailCaptionFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsVideo DescriptionVideoDetailCaption (VideoDC) (test)3.35Test Score17Speculative DecodingVideoDetailCaption ~17k visual tokens3.91Tau (τ)8Video-Language UnderstandingVideoDetailCaption0.254RG-L7