ResearchTasksUnseen UI GeneralizationUnseen UI GeneralizationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedOSExpert-EvalHuman Expert16Execution Time (Tableau, s)10Mar 10, 2026OSExpert EvalOSExpert (Explore w/ Qwen-3-VL-8B)25Tableau Success Rate8Mar 10, 2026