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

Yongkang Liu, Shi Feng, Daling Wang, Yifei Zhang


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.
Anthology ID:
2022.coling-1.54
Volume:
Proceedings of the 29th International Conference on Computational Linguistics
Month:
October
Year:
2022
Address:
Gyeongju, Republic of Korea
Venue:
COLING
SIG:
Publisher:
International Committee on Computational Linguistics
Note:
Pages:
648–659
Language:
URL:
https://aclanthology.org/2022.coling-1.54
DOI:
Bibkey:
Cite (ACL):
Yongkang Liu, Shi Feng, Daling Wang, and Yifei Zhang. 2022. MulZDG: Multilingual Code-Switching Framework for Zero-shot Dialogue Generation. In Proceedings of the 29th International Conference on Computational Linguistics, pages 648–659, Gyeongju, Republic of Korea. International Committee on Computational Linguistics.
Cite (Informal):
MulZDG: Multilingual Code-Switching Framework for Zero-shot Dialogue Generation (Liu et al., COLING 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.coling-1.54.pdf
Data
DailyDialog