MulZDG: Multilingual Code-Switching Framework for Zero-shot Dialogue Generation

TL;DR

MulZDG is a multilingual framework that enables zero-shot dialogue generation by transferring knowledge from English to other languages using code-switching datasets, improving performance without relying on large pre-trained models.

Contribution

This paper introduces MulZDG, a novel multilingual code-switching framework that facilitates zero-shot dialogue generation and enhances resource-rich language performance.

Findings

Achieves competitive zero-shot dialogue generation performance.

Effectively transfers knowledge across languages using code-switching datasets.

Improves performance of the source language with multilingual training.

Abstract

Building dialogue generation systems in a zero-shot scenario remains a huge challenge, since the typical zero-shot approaches in dialogue generation rely heavily on large-scale pre-trained language generation models such as GPT-3 and T5. The research on zero-shot dialogue generation without cumbersome language models is limited due to lacking corresponding parallel dialogue corpora. In this paper, we propose a simple but effective Multilingual learning framework for Zero-shot Dialogue Generation (dubbed as MulZDG) that can effectively transfer knowledge from an English corpus with large-scale training samples to a non-English corpus with zero samples. Besides, MulZDG can be viewed as a multilingual data augmentation method to improve the performance of the resource-rich language. First, we construct multilingual code-switching dialogue datasets via translation utterances randomly selected from monolingual English datasets. Then we employ MulZDG to train a unified multilingual dialogue model based on the code-switching datasets. The MulZDG can conduct implicit semantic alignment between different languages. Experiments on DailyDialog and DSTC7 datasets demonstrate that MulZDG not only achieve competitive performance under zero-shot case compared to training with sufficient examples but also greatly improve the performance of the source language.