ResearchTasksSemantic Safety EvaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedXS-Test CS-RLHF responsesCS-RLHF96.43XS Score3Jun 11, 2026XS Safe-RLHF responses (test)CS-RLHF91.05XS Score3Jun 11, 2026