ResearchDatasetsCategoricalHarmfulQACategoricalHarmfulQAFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsSafety EvaluationCategoricalHarmfulQA Alpaca fine-tuning (test)-22.91ASR Delta (S1-S5)42Safety EvaluationCategoricalHarmfulQA Dolly fine-tuning (test)1.27ASR (S1)21