Efficient slot correlation learning network for multi-domain dialogue state tracking

doi:10.1007/s11227-023-05217-z

CORC > 自动化研究所 > 中国科学院自动化研究所 > 精密感知与控制研究中心 > 人工智能与机器学习

	Efficient slot correlation learning network for multi-domain dialogue state tracking
	Li, Qianyu 2; Zhang, Wensheng 1,2; Huang, Mengxing 2
刊名	JOURNAL OF SUPERCOMPUTING
	2023-05-16
页码	22
关键词	Dialogue state tracking Task-oriented dialogue systems Joint goal accuracy Slot correlation
ISSN号	0920-8542
DOI	10.1007/s11227-023-05217-z
通讯作者	Zhang, Wensheng(wensheng.zhang@ia.ac.cn)
英文摘要	Task-oriented dialogue systems depend on dialogue state tracking to keep track of the intentions of users in the course of conversations. Recent studies in dialogue state tracking have achieved good performance, although the great majority of them do not consider slot correlation and just predict the value of every slot separately. In this work, we propose an efficient slot correlation learning network that can capture the correlations among slots as precisely as possible. Specifically, a BERT-base-uncased encoder is first applied to encode the dialogue context, slot names and their corresponding values. Second, we design a cross multi-head attention module to calculate and fuse attention among dialogue context embedding, slot name embedding and corresponding value embedding, which extracts relevant features and provides them to other components to fully catch the slot-specific information of every slot. Finally, a transformer encoder module is used to catch the correlations among slots. Experimental results on MultiWOZ 2.0, MultiWOZ 2.1, and MultiWOZ 2.4 datasets demonstrate the effectiveness of our approach with 55.14%, 57.22% and 76.93% joint goal accuracy, respectively, which achieves new state-of-the-art performance.
资助项目	National Key Research and Development Program of China[2018AAA0102100] ; National Natural Science Foundation of China[61976212] ; Hainan Provincial Natural Science 683 Foundation of China[621MS019]
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	SPRINGER
WOS记录号	WOS:000988684600004
资助机构	National Key Research and Development Program of China ; National Natural Science Foundation of China ; Hainan Provincial Natural Science 683 Foundation of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/53347]
专题	精密感知与控制研究中心_人工智能与机器学习
通讯作者	Zhang, Wensheng
作者单位	1.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 2.Hainan Univ, Sch Informat & Commun Engn, Haikou 570100, Hainan, Peoples R China
推荐引用方式 GB/T 7714	Li, Qianyu,Zhang, Wensheng,Huang, Mengxing. Efficient slot correlation learning network for multi-domain dialogue state tracking[J]. JOURNAL OF SUPERCOMPUTING,2023:22.
APA	Li, Qianyu,Zhang, Wensheng,&Huang, Mengxing.(2023).Efficient slot correlation learning network for multi-domain dialogue state tracking.JOURNAL OF SUPERCOMPUTING,22.
MLA	Li, Qianyu,et al."Efficient slot correlation learning network for multi-domain dialogue state tracking".JOURNAL OF SUPERCOMPUTING (2023):22.