Despite their popularity in the chatbot literature, retrieval-based models have had modest impact on task-oriented dialogue systems, with the main obstacle to their application being the low-data regime of most task-oriented dialogue tasks. Inspired by the recent success of pretraining in language modelling, we propose an effective method for deploying response selection in task-oriented dialogue. To train response selection models for task-oriented dialogue tasks, we propose a novel method which: 1) pretrains the response selection model on large general-domain conversational corpora; and then 2) fine-tunes the pretrained model for the target dialogue domain, relying only on the small in-domain dataset to capture the nuances of the given dialogue domain. Our evaluation on six diverse application domains, ranging from e-commerce to banking, demonstrates the effectiveness of the proposed training method.
@article{arxiv.1906.01543,
title = {Training Neural Response Selection for Task-Oriented Dialogue Systems},
author = {Matthew Henderson and Ivan Vulić and Daniela Gerz and Iñigo Casanueva and Paweł Budzianowski and Sam Coope and Georgios Spithourakis and Tsung-Hsien Wen and Nikola Mrkšić and Pei-Hao Su},
journal= {arXiv preprint arXiv:1906.01543},
year = {2019}
}