Self-Training with Purpose Preserving Augmentation Improves Few-shot Generative Dialogue State Tracking

TL;DR

This paper introduces a self-training framework with Purpose Preserving Augmentation to improve few-shot generative dialogue state tracking, reducing labeling effort and enhancing performance on benchmark datasets.

Contribution

It presents a novel self-training approach combined with PPAug for few-shot DST, addressing overfitting and improving accuracy with limited labeled data.

Findings

Achieved 4% performance increase on MultiWOZ 2.1 with 10% labeled data

Enhanced slot-recall by 8.34% for unseen values

Demonstrated effectiveness of PPAug in preventing overfitting

Abstract

In dialogue state tracking (DST), labeling the dataset involves considerable human labor. We propose a new self-training framework for few-shot generative DST that utilize unlabeled data. Our self-training method iteratively improves the model by pseudo labeling and employs Purpose Preserving Augmentation (PPAug) to prevent overfitting. We increaese the few-shot 10% performance by approximately 4% on MultiWOZ 2.1 and enhances the slot-recall 8.34% for unseen values compared to baseline.