ResearchDatasetsTS-BenchFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsSafety detectionTS-Bench AgentDojo-Traj (eval)1.36Efficiency (s/sample)4Safety detectionTS-Bench AgentHarm-Traj (eval)1.36Latency (s/sample)4Safety risk detectionTS-Bench86Overall F1 Score3