ResearchTasksVLM ReasoningVLM ReasoningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedBlackJack (In-Distribution)GFlowVLM w/ DB43.8Success Rate15Feb 26, 2026NumberLine Out-of-DistributionGFlowVLM w/ DB18.6Success Rate15Feb 26, 2026NumberLine (In-Distribution)GFlowVLM w/ Var-TB100Success Rate15Feb 26, 2026