Towards Task-Oriented Dialogue in Mixed Domains

TL;DR

This paper explores mixed-domain task-oriented dialogue systems, demonstrating the importance of specialized state tracking and proposing a hybrid system that significantly improves belief tracking accuracy in multi-domain dialogues.

Contribution

It introduces a hybrid dialogue system that enhances belief tracking accuracy by 28%, highlighting the importance of specialized state tracking in multi-domain settings.

Findings

Specialized state tracking outperforms end-to-end systems.

Hybrid system improves belief tracking accuracy by 28%.

Insights for enhancing commercial chatbot platforms.

Abstract

This work investigates the task-oriented dialogue problem in mixed-domain settings. We study the effect of alternating between different domains in sequences of dialogue turns using two related state-of-the-art dialogue systems. We first show that a specialized state tracking component in multiple domains plays an important role and gives better results than an end-to-end task-oriented dialogue system. We then propose a hybrid system which is able to improve the belief tracking accuracy of about 28% of average absolute point on a standard multi-domain dialogue dataset. These experimental results give some useful insights for improving our commercial chatbot platform FPT.AI, which is currently deployed for many practical chatbot applications.