ResearchTasksAnnotationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedAnnotation Task Suite Toxicity and Safety Domains (various)GPT-4o-mini84.4Accuracy72Jun 2, 2026EC PFMBench (test)ProstT576.829Score2Feb 26, 2026