ResearchDatasetsSeeTRUESeeTRUEFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsVisual faithfulness evaluationSeeTRUE82F1 Score15Image-Text AlignmentSeeTRUE63.5Winoground14Binary ClassificationSeetrue (test)83.67Macro F1 Score5