ResearchDatasetsOrBenchFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsOverrefusal EvaluationOrBench-H99.85RR32Safety Boundary Over-RefusalORBench7.2ORBench Over-Refusal Rate20Safety and Informativeness EvaluationOrBench Hard75Deception Rate (Safe)4Safety EvaluationOrBench Toxic81.7Safety Score4