A Multi-Task BERT Model for Schema-Guided Dialogue State Tracking
This work addresses the need for more efficient and robust dialogue state tracking in task-oriented systems, offering a practical improvement over existing methods.
The paper tackles the problem of improving computational efficiency and performance in schema-guided dialogue state tracking by proposing a single multi-task BERT model that jointly handles intent prediction, requested slot prediction, and slot filling, with an efficient encoding method. Evaluation on the SGD dataset shows it outperforms the baseline SGP-DST by a large margin and performs competitively with state-of-the-art while being significantly more efficient.
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.