Scalable Neural Dialogue State Tracking

TL;DR

This paper introduces G-SAT, a neural dialogue state tracker that achieves high accuracy with significantly reduced latency, enabling real-time deployment in multilingual dialogue systems.

Contribution

The paper presents G-SAT, a novel neural model for dialogue state tracking that is faster and maintains high performance across multiple languages.

Findings

G-SAT is over 15 times faster than existing systems.

G-SAT achieves competitive accuracy on multilingual datasets.

The model scales well with the number of dialogue slots.

Abstract

A Dialogue State Tracker (DST) is a key component in a dialogue system aiming at estimating the beliefs of possible user goals at each dialogue turn. Most of the current DST trackers make use of recurrent neural networks and are based on complex architectures that manage several aspects of a dialogue, including the user utterance, the system actions, and the slot-value pairs defined in a domain ontology. However, the complexity of such neural architectures incurs into a considerable latency in the dialogue state prediction, which limits the deployments of the models in real-world applications, particularly when task scalability (i.e. amount of slots) is a crucial factor. In this paper, we propose an innovative neural model for dialogue state tracking, named Global encoder and Slot-Attentive decoders (G-SAT), which can predict the dialogue state with a very low latency time, while maintaining high-level performance. We report experiments on three different languages (English, Italian, and German) of the WoZ2.0 dataset, and show that the proposed approach provides competitive advantages over state-of-art DST systems, both in terms of accuracy and in terms of time complexity for predictions, being over 15 times faster than the other systems.