Proceedings of the First Workshop on NLP for Conversational AI

Progress in Machine Learning is often driven by the availability of large datasets, and consistent evaluation metrics for comparing modeling approaches.To this end, we present a repository of conversational datasets consisting of hundreds of millions of examples, and a standardised evaluation procedure for conversational response selection models using 1-of-100 accuracy.The repository contains scripts that allow researchers to reproduce the standard datasets, or to adapt the pre-processing and data filtering steps to their needs.We introduce and evaluate several competitive baselines for conversational response selection, whose implementations are shared in the repository, as well as a neural encoder model that is trained on the entire training set.Built from Training size Testing size Reddit 3.7 billion comments in threaded conversations 654,396,778 72,616,937 OpenSubtitles over 400 million lines from movie and television subtitles (also available in other languages)