ResearchDatasetsCOLDCOLDFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsToxicity ClassificationCOLD (test)94.33Accuracy19Content SafetyCOLD (test)74.8Macro F1-Score8Toxicity DetectionCOLD75.3ASR7