Discovering Dialog Structure Graph for Coherent Dialog Generation

Jun Xu, Zeyang Lei, Haifeng Wang, Zheng-Yu Niu, Hua Wu, Wanxiang Che


Abstract
Learning discrete dialog structure graph from human-human dialogs yields basic insights into the structure of conversation, and also provides background knowledge to facilitate dialog generation. However, this problem is less studied in open-domain dialogue. In this paper, we conduct unsupervised discovery of discrete dialog structure from chitchat corpora, and then leverage it to facilitate coherent dialog generation in downstream systems. To this end, we present an unsupervised model, Discrete Variational Auto-Encoder with Graph Neural Network (DVAE-GNN), to discover discrete hierarchical latent dialog states (at the level of both session and utterance) and their transitions from corpus as a dialog structure graph. Then we leverage it as background knowledge to facilitate dialog management in a RL based dialog system. Experimental results on two benchmark corpora confirm that DVAE-GNN can discover meaningful dialog structure graph, and the use of dialog structure as background knowledge can significantly improve multi-turn coherence.
Anthology ID:
2021.acl-long.136
Volume:
Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
Month:
August
Year:
2021
Address:
Online
Editors:
Chengqing Zong, Fei Xia, Wenjie Li, Roberto Navigli
Venues:
ACL | IJCNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1726–1739
Language:
URL:
https://aclanthology.org/2021.acl-long.136
DOI:
10.18653/v1/2021.acl-long.136
Bibkey:
Cite (ACL):
Jun Xu, Zeyang Lei, Haifeng Wang, Zheng-Yu Niu, Hua Wu, and Wanxiang Che. 2021. Discovering Dialog Structure Graph for Coherent Dialog Generation. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 1726–1739, Online. Association for Computational Linguistics.
Cite (Informal):
Discovering Dialog Structure Graph for Coherent Dialog Generation (Xu et al., ACL-IJCNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.acl-long.136.pdf
Video:
 https://aclanthology.org/2021.acl-long.136.mp4
Data
Douban