ResearchTasksOpen-domain dialogue red teamingFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedConvAI2 filtered (test)BRT (e+r)16.9RSR7Feb 26, 2026Bloom ZS (filtered) (test)BRT (e+r)16.3RSR7Feb 26, 2026