ResearchTasksVLM-as-Judge EvaluationVLM-as-Judge EvaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedArcBench VLM-as-Judge ClosedHTML58.5TQ15Apr 15, 2026ArcBench VLM-as-Judge OpenHTML86.11TQ15Apr 15, 2026