ResearchTasksToxic Language SuppressionFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedRealToxicityPrompts 10K nontoxic prompts GPT2-large generation (test)LLMBRACES0.172Max Toxicity7Feb 26, 2026
RealToxicityPrompts 10K nontoxic prompts GPT2-large generation (test)LLMBRACES0.172Max Toxicity7Feb 26, 2026