A Multi-Task BERT Model for Schema-Guided Dialogue State Tracking

TL;DR

This paper introduces a multi-task BERT model for dialogue state tracking that efficiently handles intent prediction, slot filling, and requested slots, outperforming existing models while reducing computational complexity.

Contribution

A novel single multi-task BERT-based model that jointly addresses multiple DST tasks with an efficient encoding scheme, improving performance and efficiency.

Findings

Outperforms baseline SGP-DST significantly

Achieves competitive results with state-of-the-art models

More computationally efficient than previous approaches

Abstract

Task-oriented dialogue systems often employ a Dialogue State Tracker (DST) to successfully complete conversations. Recent state-of-the-art DST implementations rely on schemata of diverse services to improve model robustness and handle zero-shot generalization to new domains [1], however such methods [2, 3] typically require multiple large scale transformer models and long input sequences to perform well. We propose a single multi-task BERT-based model that jointly solves the three DST tasks of intent prediction, requested slot prediction and slot filling. Moreover, we propose an efficient and parsimonious encoding of the dialogue history and service schemata that is shown to further improve performance. Evaluation on the SGD dataset shows that our approach outperforms the baseline SGP-DST by a large margin and performs well compared to the state-of-the-art, while being significantly more computationally efficient. Extensive ablation studies are performed to examine the contributing factors to the success of our model.