ResearchTasksIn-the-wild model generalizationIn-the-wild model generalizationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedHuman Bench AverageQwen3VL-2B57.9NSE Score14Feb 26, 2026Human Bench Text-based DemoQwen3VL-32B23.4NSE14Feb 26, 2026Human Bench Vision-based DemoProgressLM-RL-3B15.5NSE14Feb 26, 2026