ResearchDatasetsPrompt-ManipulationPrompt-ManipulationFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsHidden Objective MonitoringPrompt-Manipulation A1 Monitor Notice100Exit Detection Rate22Hidden Objective MonitoringPrompt-Manipulation A1 Standard100Exit Success Rate22