ResearchBenchmarksTask-oriented Dialog Response Generation on MultiWOZ human evaluation (test)Follow3.65DiversityDAMD2.07962.48732.8953.3027Nov 24, 2019Evaluation ResultsMethodMethodLinksDiversityAppropriatenessGood RateOK RateInvalid RateDAMDData Augmentation=trueData Augmentation=true2019.113.652.536327.19.9DAMDData Augmentation=falseData Augmentation=false2019.113.122.556.537.46.1HDSAData Augmentation=trueData Augmentation=true2019.112.142.4757.532.510