Effective Sequence-to-Sequence Dialogue State Tracking

Jeffrey Zhao, Mahdis Mahdieh, Ye Zhang, Yuan Cao, Yonghui Wu


Abstract
Sequence-to-sequence models have been applied to a wide variety of NLP tasks, but how to properly use them for dialogue state tracking has not been systematically investigated. In this paper, we study this problem from the perspectives of pre-training objectives as well as the formats of context representations. We demonstrate that the choice of pre-training objective makes a significant difference to the state tracking quality. In particular, we find that masked span prediction is more effective than auto-regressive language modeling. We also explore using Pegasus, a span prediction-based pre-training objective for text summarization, for the state tracking model. We found that pre-training for the seemingly distant summarization task works surprisingly well for dialogue state tracking. In addition, we found that while recurrent state context representation works also reasonably well, the model may have a hard time recovering from earlier mistakes. We conducted experiments on the MultiWOZ 2.1-2.4, WOZ 2.0, and DSTC2 datasets with consistent observations.
Anthology ID:
2021.emnlp-main.593
Volume:
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2021
Address:
Online and Punta Cana, Dominican Republic
Editors:
Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
7486–7493
Language:
URL:
https://aclanthology.org/2021.emnlp-main.593
DOI:
10.18653/v1/2021.emnlp-main.593
Bibkey:
Cite (ACL):
Jeffrey Zhao, Mahdis Mahdieh, Ye Zhang, Yuan Cao, and Yonghui Wu. 2021. Effective Sequence-to-Sequence Dialogue State Tracking. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 7486–7493, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
Effective Sequence-to-Sequence Dialogue State Tracking (Zhao et al., EMNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.emnlp-main.593.pdf
Video:
 https://aclanthology.org/2021.emnlp-main.593.mp4
Code
 smartyfh/MultiWOZ2.4
Data
Dialogue State Tracking ChallengeMultiWOZWizard-of-Oz