ResearchDatasetsToolEmuFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsAgent Behavioral Safety and Helpfulness EvaluationToolEmu97.9Safety Rate42Agent Safety EvaluationToolEmu99Safety36Policy-refinementToolEmu33.1IOR16Graph-based Agent Memory PoisoningToolEmu97Utilization (Util.)5Propagation DetectionToolEmu filtered 79-case injection-like62.3Precision2