ResearchDatasetsAgentDynFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsPrompt Injection DetectionAgentDyn83.1Source Precision30Dynamic Agent Security and Utility EvaluationAgentDyn52ASR22Prompt Injection DiscoveryAgentDyn78Source Diversity14Prompt Injection DefenseAgentDyn1ASR9Agent Safety EvaluationAgentDyn100Benign Rate2