Data used for ACL 2020 paper “None of the Above”:Measure Uncertainty in Dialog Response Retrieval
You can download our preprocessed data, together with our results, from here
model.py and train_retrieval.py are code of the LSTM model, and train_nota.py is code of the LogReg model