ResearchTasksCorrelation to Human PreferenceFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedHuman Preference Annotations Physics Scenarios (test)Qwen2.5 VL39.4Correlation: Ball Collision4Mar 9, 2026
Human Preference Annotations Physics Scenarios (test)Qwen2.5 VL39.4Correlation: Ball Collision4Mar 9, 2026