Pretrain-Finetune Based Training of Task-Oriented Dialogue Systems in a Real-World Setting

Abstract

One main challenge in building task-oriented dialogue systems is the limited amount of supervised training data available. In this work, we present a method for training retrieval-based dialogue systems using a small amount of high-quality, annotated data and a larger, unlabeled dataset. We show that pretraining using unlabeled data can bring better model performance with a 31% boost in Recall@1 compared with no pretraining. The proposed finetuning technique based on a small amount of high-quality, annotated data resulted in 26% offline and 33% online performance improvement in Recall@1 over the pretrained model. The model is deployed in an agent-support application and evaluated on live customer service contacts, providing additional insights into the real-world implications compared with most other publications in the domain often using asynchronous transcripts (e.g. Reddit data). The high performance of 74% Recall@1 shown in the customer service example demonstrates the effectiveness of this pretrain-finetune approach in dealing with the limited supervised data challenge.