ResearchTasksVisual Reasoning RobustnessVisual Reasoning RobustnessFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedVisArgs specific masking 1.0 (test)Gemini-1.5-Flash58.78Answer Flip Rate4Feb 26, 2026