跳到主要导航 跳到搜索 跳到主要内容

SLDT: Sequential Latent Document Transformer for Multilingual Document-based Dialogue

  • Zhanyu Ma
  • , Zeming Liu
  • , Jian Ye*
  • *此作品的通讯作者
  • CAS - Institute of Computing Technology
  • University of Chinese Academy of Sciences
  • Beijing Key Lab. of Mobile Computing and Pervasive Device
  • Harbin Institute of Technology

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

Multilingual document-grounded dialogue, where the system is required to generate responses based on both the conversation multilingual context and external knowledge sources. Traditional pipeline methods for knowledge identification and response generation, while effective in certain scenarios, suffer from error propagation issues and fail to capture the interdependence between these two sub-tasks. To overcome these challenges, we propose the application of the SLDT method, which treats passage-knowledge selection as a sequential decision process rather than a single-step decision process. We achieved the winner 3rd in dialdoc 2023 and we also validated the effectiveness of our method on other datasets. The ablation experiment also shows that our method significantly improves the basic model compared to other methods.

源语言英语
主期刊名DialDoc 2023 - Proceedings of the 3rd DialDoc Workshop on Document-Grounded Dialogue and Conversational Question Answering, Proceedings of the Workshop
编辑Smaranda Muresan, Vivian Chen, Casey Kennington, David Vandyke, Nina Dethlefs, Koji Inoue, Erik Ekstedt, Stefan Ultes
出版商Association for Computational Linguistics (ACL)
57-67
页数11
ISBN(电子版)9781959429982
出版状态已出版 - 2023
已对外发布
活动3rd Workshop on Document-grounded Dialogue and Conversational Question Answering, DialDoc 2023, co-located with ACL 2023 - Toronto, 加拿大
期限: 13 7月 2023 → …

出版系列

姓名Proceedings of the Annual Meeting of the Association for Computational Linguistics
ISSN(印刷版)0736-587X

会议

会议3rd Workshop on Document-grounded Dialogue and Conversational Question Answering, DialDoc 2023, co-located with ACL 2023
国家/地区加拿大
Toronto
时期13/07/23 → …

学术指纹

探究 'SLDT: Sequential Latent Document Transformer for Multilingual Document-based Dialogue' 的科研主题。它们共同构成独一无二的学术指纹。

引用此