ResearchDatasetsToDs benchmarkToDs benchmarkFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsNatural Language GenerationToDs benchmark GPT-2 backbone (test)3.42EER11Dialogue State TrackingToDs benchmark GPT-2 backbone (test)50.03JGA11