Transformer-based Multi-Party Conversation Generation using Dialogue Discourse Acts Planning

Alexander Chernyavskiy, Dmitry Ilvovsky


Abstract
Recent transformer-based approaches to multi-party conversation generation may produce syntactically coherent but discursively inconsistent dialogues in some cases. To address this issue, we propose an approach to integrate a dialogue act planning stage into the end-to-end transformer-based generation pipeline. This approach consists of a transformer fine-tuning procedure based on linearized dialogue representations that include special discourse tokens. The obtained results demonstrate that incorporating discourse tokens into training sequences is sufficient to significantly improve dialogue consistency and overall generation quality. The suggested approach performs well, including for automatically annotated data. Apart from that, it is observed that increasing the weight of the discourse planning task in the loss function accelerates learning convergence.
Anthology ID:
2023.sigdial-1.48
Volume:
Proceedings of the 24th Annual Meeting of the Special Interest Group on Discourse and Dialogue
Month:
September
Year:
2023
Address:
Prague, Czechia
Editors:
Svetlana Stoyanchev, Shafiq Joty, David Schlangen, Ondrej Dusek, Casey Kennington, Malihe Alikhani
Venue:
SIGDIAL
SIG:
SIGDIAL
Publisher:
Association for Computational Linguistics
Note:
Pages:
519–529
Language:
URL:
https://aclanthology.org/2023.sigdial-1.48
DOI:
10.18653/v1/2023.sigdial-1.48
Bibkey:
Cite (ACL):
Alexander Chernyavskiy and Dmitry Ilvovsky. 2023. Transformer-based Multi-Party Conversation Generation using Dialogue Discourse Acts Planning. In Proceedings of the 24th Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 519–529, Prague, Czechia. Association for Computational Linguistics.
Cite (Informal):
Transformer-based Multi-Party Conversation Generation using Dialogue Discourse Acts Planning (Chernyavskiy & Ilvovsky, SIGDIAL 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.sigdial-1.48.pdf