Selecting Stickers in Open-Domain Dialogue through Multitask Learning

Zhexin Zhang, Yeshuang Zhu, Zhengcong Fei, Jinchao Zhang, Jie Zhou


Abstract
With the increasing popularity of online chatting, stickers are becoming important in our online communication. Selecting appropriate stickers in open-domain dialogue requires a comprehensive understanding of both dialogues and stickers, as well as the relationship between the two types of modalities. To tackle these challenges, we propose a multitask learning method comprised of three auxiliary tasks to enhance the understanding of dialogue history, emotion and semantic meaning of stickers. Extensive experiments conducted on a recent challenging dataset show that our model can better combine the multimodal information and achieve significantly higher accuracy over strong baselines. Ablation study further verifies the effectiveness of each auxiliary task. Our code is available at https://github.com/nonstopfor/Sticker-Selection.
Anthology ID:
2022.findings-acl.241
Volume:
Findings of the Association for Computational Linguistics: ACL 2022
Month:
May
Year:
2022
Address:
Dublin, Ireland
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3053–3060
Language:
URL:
https://aclanthology.org/2022.findings-acl.241
DOI:
10.18653/v1/2022.findings-acl.241
Bibkey:
Cite (ACL):
Zhexin Zhang, Yeshuang Zhu, Zhengcong Fei, Jinchao Zhang, and Jie Zhou. 2022. Selecting Stickers in Open-Domain Dialogue through Multitask Learning. In Findings of the Association for Computational Linguistics: ACL 2022, pages 3053–3060, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
Selecting Stickers in Open-Domain Dialogue through Multitask Learning (Zhang et al., Findings 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2022.findings-acl.241.pdf
Software:
 2022.findings-acl.241.software.zip
Code
 nonstopfor/sticker-selection