Robust Cross-lingual Task-oriented Dialogue

doi:10.1145/3457571

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 自然语言处理

	Robust Cross-lingual Task-oriented Dialogue
	Xiang, Lu 1,2; Zhu, Junnan 1,2; Zhao, Yang 1,2; Zhou, Yu 1,2; Zong, Chengqing 1,2
刊名	ACM TRANSACTIONS ON ASIAN AND LOW-RESOURCE LANGUAGE INFORMATION PROCESSING
	2021-11-01
卷号	20 期号:6 页码:24
关键词	Cross-lingual dialogue system adversarial learning knowledge robustness
ISSN号	2375-4699
DOI	10.1145/3457571
通讯作者	Xiang, Lu(lu.xiang@nlpr.ia.ac.cn)
英文摘要	Cross-lingual dialogue systems are increasingly important in e-commerce and customer service due to the rapid progress of globalization. In real-world system deployment, machine translation (MT) services are often used before and after the dialogue system to bridge different languages. However, noises and errors introduced in the MT process will result in the dialogue system's low robustness, making the system's performance far from satisfactory. In this article, we propose a novel MT-oriented noise enhanced framework that exploits multi-granularityMTnoises and injects such noises into the dialogue system to improve the dialogue system's robustness. Specifically, we first design a method to automatically construct multi-granularity MT-oriented noises and multi-granularity adversarial examples, which contain abundant noise knowledge oriented to MT. Then, we propose two strategies to incorporate the noise knowledge: (i) Utterance-level adversarial learning and (ii) Knowledge-level guided method. The former adopts adversarial learning to learn a perturbation-invariant encoder, guiding the dialogue system to learn noise-independent hidden representations. The latter explicitly incorporates the multi-granularity noises, which contain the noise tokens and their possible correct forms, into the training and inference process, thus improving the dialogue system's robustness. Experimental results on three dialoguemodels, two dialogue datasets, and two language pairs have shown that the proposed framework significantly improves the performance of the cross-lingual dialogue system.
资助项目	National Key Research and Development Program of China[2017YFB1002103]
WOS关键词	SPOKEN ; NETWORKS
WOS研究方向	Computer Science
语种	英语
出版者	ASSOC COMPUTING MACHINERY
WOS记录号	WOS:000721586800002
资助机构	National Key Research and Development Program of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/46443]
专题	模式识别国家重点实验室_自然语言处理
通讯作者	Xiang, Lu
作者单位	1.Intelligence Bldg,95 Zhongguancun East Rd, Beijing 100190, Peoples R China 2.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Sch Artificial Intelligence,Univ Chinese Acad Sci, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Xiang, Lu,Zhu, Junnan,Zhao, Yang,et al. Robust Cross-lingual Task-oriented Dialogue[J]. ACM TRANSACTIONS ON ASIAN AND LOW-RESOURCE LANGUAGE INFORMATION PROCESSING,2021,20(6):24.
APA	Xiang, Lu,Zhu, Junnan,Zhao, Yang,Zhou, Yu,&Zong, Chengqing.(2021).Robust Cross-lingual Task-oriented Dialogue.ACM TRANSACTIONS ON ASIAN AND LOW-RESOURCE LANGUAGE INFORMATION PROCESSING,20(6),24.
MLA	Xiang, Lu,et al."Robust Cross-lingual Task-oriented Dialogue".ACM TRANSACTIONS ON ASIAN AND LOW-RESOURCE LANGUAGE INFORMATION PROCESSING 20.6(2021):24.