ResearchTasksVision-Language CaptioningVision-Language CaptioningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedxBD (test)QwenVL77.86CLIPScore (%)8Feb 26, 2026RescueNet (test)QwenVL99.78CLIPScore8Feb 26, 2026