ResearchDatasetsOOD Evaluation SuiteOOD Evaluation SuiteFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsGeneralization across real-world cross-view spatial reasoning benchmarksOOD Evaluation Suite Aggregate40Avg OOD Performance13Click-in-targetOOD Evaluation Suite (ScreenSpot-Pro, OSWorld-G, ScreenSpot-V2) unweighted mean60Click-in-Target Accuracy (SS-Pro)11
Generalization across real-world cross-view spatial reasoning benchmarksOOD Evaluation Suite Aggregate40Avg OOD Performance13
Click-in-targetOOD Evaluation Suite (ScreenSpot-Pro, OSWorld-G, ScreenSpot-V2) unweighted mean60Click-in-Target Accuracy (SS-Pro)11