Publication Type
Conference Proceeding Article
Version
publishedVersion
Publication Date
6-2024
Abstract
Mixed initiative serves as one of the key factors in controlling conversation directions. For a speaker, responding passively or leading proactively would result in rather different responses. However, most dialogue systems focus on training a holistic response generation model without any distinction among different initiatives. It leads to the cross-contamination problem, where the model confuses different initiatives and generates inappropriate responses. Moreover, obtaining plenty of human annotations for initiative labels can be expensive. To address this issue, we propose a general mix-Initiative Dynamic Prefix Tuning framework (IDPT) to decouple different initiatives from the generation model, which learns initiative-aware prefixes in both supervised and unsupervised settings. Specifically, IDPT decouples initiative factors into different prefix parameters and uses the attention mechanism to adjust the selection of initiatives in guiding generation dynamically. The prefix parameters can be tuned towards accurate initiative prediction as well as mix-initiative response generation. Extensive experiments on two public dialogue datasets show that the proposed IDPT outperforms previous baselines on both automatic metrics and human evaluations. It also manages to generate appropriate responses with manipulated initiatives.
Keywords
Dynamic prefix tuning framework, Response generation model, Dialogue systems
Discipline
Artificial Intelligence and Robotics | Computer Sciences
Research Areas
Data Science and Engineering; Intelligent Systems and Optimization
Areas of Excellence
Digital transformation
Publication
Proceedings of the Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2024) : Mexico City, Mexico, June 16-21
Volume
1
First Page
8748
Last Page
8761
Identifier
10.18653/v1/2024.naacl-long.485
Publisher
Association for Computational Linguistics
City or Country
Mexico City, Mexico
Citation
NIE, Yuxiang; HUANG, Heyan; MAO, Xian-Ling; and LIAO, Lizi.
Mix-initiative response generation with dynamic prefix tuning. (2024). Proceedings of the Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2024) : Mexico City, Mexico, June 16-21. 1, 8748-8761.
Available at: https://ink.library.smu.edu.sg/sis_research/9701
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Additional URL
https://doi.org/10.18653/v1/2024.naacl-long.485