Scalable Multi-Domain Dialogue State Tracking

TL;DR

This paper introduces a scalable, slot-independent deep learning framework for dialogue state tracking that efficiently handles unbounded and dynamic slot values, enabling quick adaptation to new domains.

Contribution

The paper proposes a novel, slot-independent DST approach that uses candidate sets from dialogue history, improving scalability and transfer learning for multi-domain dialogue systems.

Findings

Effective handling of unbounded slot values

Improved scalability with bounded candidate sets

Facilitates quick domain adaptation

Abstract

Dialogue state tracking (DST) is a key component of task-oriented dialogue systems. DST estimates the user's goal at each user turn given the interaction until then. State of the art approaches for state tracking rely on deep learning methods, and represent dialogue state as a distribution over all possible slot values for each slot present in the ontology. Such a representation is not scalable when the set of possible values are unbounded (e.g., date, time or location) or dynamic (e.g., movies or usernames). Furthermore, training of such models requires labeled data, where each user turn is annotated with the dialogue state, which makes building models for new domains challenging. In this paper, we present a scalable multi-domain deep learning based approach for DST. We introduce a novel framework for state tracking which is independent of the slot value set, and represent the dialogue state as a distribution over a set of values of interest (candidate set) derived from the dialogue history or knowledge. Restricting these candidate sets to be bounded in size addresses the problem of slot-scalability. Furthermore, by leveraging the slot-independent architecture and transfer learning, we show that our proposed approach facilitates quick adaptation to new domains.