ResearchDatasetsVLNVLNFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsDense Video Object CaptioningVLN (val)23.4CapA4