ResearchDatasetsMedVidQAMedVidQAFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsInstructional Temporal Answer Grounding in VideoMedVidQA (test)80.65R@0.319Long-context ProcessingMedVidQA with subtitles65.8mIoU2Long-context ProcessingMedVidQA43.4mIoU2