ResearchTasksDialogue CoherenceFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedBMELD (val)Human Reference61.03Precision (1st)7Feb 26, 2026De-En Base (test)Human Reference0.6661st Precision7Feb 26, 2026DailyDialogDD++3.24QuantiDCE6Feb 26, 2026Dialogue Coherence EvaluationGPT-4o4.19Anger Score5Feb 26, 2026PersonaChatDialoGPS3.03QuantiDCE5Feb 26, 2026