Plan, Generate and Complicate: Improving Low-resource Dialogue State Tracking via Easy-to-Difficult Zero-shot Data Augmentation
This work addresses data scarcity for dialogue state tracking, particularly in low-resource settings, by introducing a novel data augmentation method that enhances model capability for co-reference slot tracking and mitigates output order issues.
The paper tackles the problem of low-resource dialogue state tracking by proposing an easy-to-difficult zero-shot data augmentation framework that uses large language models to generate and complicate dialogue data, resulting in improved performance on the MultiWOZ benchmark compared to previous baselines.
Data augmentation methods have been a promising direction to improve the performance of small models for low-resource dialogue state tracking. However, traditional methods rely on pre-defined user goals and neglect the importance of data complexity in this task. In this paper, we propose EDZ-DA, an Easy-to-Difficult Zero-shot Data Augmentation framework for low-resource dialogue state tracking that utilizes large language models to automatically catch the relationships of different domains and then generate the dialogue data. We also complicate the dialogues based on the domain relation to enhance the model's capability for co-reference slot tracking. Furthermore, we permute slot values to mitigate the influence of output orders and the problem of incomplete value generation. Experimental results illustrate the superiority of our proposed method compared to previous strong data augmentation baselines on MultiWOZ.