ResearchDatasetsOSFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsOperating System Interaction Agent TaskOS53.6Success Rate19Sentence ClassificationOS full (test)95.1Accuracy9Traffic Speed PredictionOS I45.86RMSE8Reasoning-Level Denial-of-ServiceOS Environment Injection (test)80E2E Success4ClusteringOS (test)77.49Accuracy2