Semi-Supervised Knowledge-Grounded Pre-training for Task-Oriented Dialog Systems
This work addresses the practical issue of limited labeled data for task-oriented dialog systems, though it appears incremental as it builds on existing neural approaches and focuses on a specific dataset.
The paper tackles the problem of costly manual labeling for task-oriented dialog systems by proposing a semi-supervised knowledge-grounded pre-training approach, achieving first place in a challenge with improvements of +7.64 BLEU and +13.6% Success over the second place.
Recent advances in neural approaches greatly improve task-oriented dialogue (TOD) systems which assist users to accomplish their goals. However, such systems rely on costly manually labeled dialogs which are not available in practical scenarios. In this paper, we present our models for Track 2 of the SereTOD 2022 challenge, which is the first challenge of building semi-supervised and reinforced TOD systems on a large-scale real-world Chinese TOD dataset MobileCS. We build a knowledge-grounded dialog model to formulate dialog history and local KB as input and predict the system response. And we perform semi-supervised pre-training both on the labeled and unlabeled data. Our system achieves the first place both in the automatic evaluation and human interaction, especially with higher BLEU (+7.64) and Success (+13.6\%) than the second place.