ResearchTasksContent Safety ClassificationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedStatic Content Safety Evaluation SetsYuvion-8B94.3Adv. & Traffic9Jun 29, 2026XsTestGradient-Controlled Decoding (GCD)92.38Precision7Apr 8, 2026ToxicChatGradSafe75.46Precision6Apr 8, 2026