Dialogue Distillation: Open-Domain Dialogue Augmentation Using Unpaired Data

TL;DR

This paper introduces a novel data augmentation method for open-domain dialogue systems that leverages unpaired data through dialogue and model distillation, improving model performance and dialogue quality.

Contribution

It proposes a combined data-level and model-level distillation approach to effectively utilize unpaired data for dialogue augmentation, which is a novel contribution.

Findings

Enhanced dialogue quality with diverse content

Improved performance of baseline dialogue models

Effective filtering of low-quality augmented dialogues

Abstract

Recent advances in open-domain dialogue systems rely on the success of neural models that are trained on large-scale data. However, collecting large-scale dialogue data is usually time-consuming and labor-intensive. To address this data dilemma, we propose a novel data augmentation method for training open-domain dialogue models by utilizing unpaired data. Specifically, a data-level distillation process is first proposed to construct augmented dialogues where both post and response are retrieved from the unpaired data. A ranking module is employed to filter out low-quality dialogues. Further, a model-level distillation process is employed to distill a teacher model trained on high-quality paired data to augmented dialogue pairs, thereby preventing dialogue models from being affected by the noise in the augmented data. Automatic and manual evaluation indicates that our method can produce high-quality dialogue pairs with diverse contents, and the proposed data-level and model-level dialogue distillation can improve the performance of competitive baselines.