ResearchDatasetsSafety Elicitation DatasetFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsSafety Elicitation via Prompt and Query RefinementSafety Elicitation Dataset gpt-5-mini judge4.61Response Realism Score6
Safety Elicitation via Prompt and Query RefinementSafety Elicitation Dataset gpt-5-mini judge4.61Response Realism Score6