ResearchDatasetsDeceptionBenchFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsDeception DetectionDeceptionBench (full 600-sample)97.4Response AUROC15Deception DetectionDeceptionBench93.5Response AUROC6Instruction Signal RecoveryDeceptionBench68.6Judge Score (R)4