UniGDD: A Unified Generative Framework for Goal-Oriented Document-Grounded Dialogue

Chang Gao, Wenxuan Zhang, Wai Lam


Abstract
The goal-oriented document-grounded dialogue aims at responding to the user query based on the dialogue context and supporting document. Existing studies tackle this problem by decomposing it into two sub-tasks: knowledge identification and response generation. However, such pipeline methods would unavoidably suffer from the error propagation issue. This paper proposes to unify these two sub-tasks via sequentially generating the grounding knowledge and the response. We further develop a prompt-connected multi-task learning strategy to model the characteristics and connections of different tasks and introduce linear temperature scheduling to reduce the negative effect of irrelevant document information. Experimental results demonstrate the effectiveness of our framework.
Anthology ID:
2022.acl-short.66
Volume:
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
Month:
May
Year:
2022
Address:
Dublin, Ireland
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
599–605
Language:
URL:
https://aclanthology.org/2022.acl-short.66
DOI:
10.18653/v1/2022.acl-short.66
Bibkey:
Cite (ACL):
Chang Gao, Wenxuan Zhang, and Wai Lam. 2022. UniGDD: A Unified Generative Framework for Goal-Oriented Document-Grounded Dialogue. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 599–605, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
UniGDD: A Unified Generative Framework for Goal-Oriented Document-Grounded Dialogue (Gao et al., ACL 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.acl-short.66.pdf
Code
 gao-xiao-bai/UniGDD
Data
Doc2Dial