ResearchBenchmarksDialog Response Generation on FEWSHOTWOZ train 1.0 (test)Follow17.21BLEUSC-GPT5.63488.639911.64514.6501Feb 27, 2020Evaluation ResultsMethodMethodLinksBLEUERRSC-GPTbackbone=GPT2-Medium,...backbone=GPT2-Medium, number_of_parameters=345M2020.0217.217.74GPT-2backbone=GPT2-Medium,...backbone=GPT2-Medium, number_of_parameters=345M2020.0213.7219.26SC-LSTMmodel_type=Semanticall...model_type=Semantically Conditioned LSTM2020.026.08189.88