ResearchTasksVideo-based Multimodal UnderstandingFor AI assistantsMarkdownContent indexVideo-based Multimodal UnderstandingFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedSEED-Bench videoQwen2.5VL61.3Accuracy8Feb 26, 2026