Fact-Driven Abstractive Summarization by Utilizing Multi-Granular Multi-Relational Knowledge

doi:10.1109/TASLP.2022.3161157

CASIA OpenIR > 多模态人工智能系统全国重点实验室 > 自然语言处理

	Fact-Driven Abstractive Summarization by Utilizing Multi-Granular Multi-Relational Knowledge
	Mao, Qianren 1,2; Li, Jianxin 1,2; Peng, Hao 1,2; He, Shizhu3,4 ; Wang, Lihong 5; Yu, Philip S.6; Wang, Zheng 7
发表期刊	IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING
ISSN	2329-9290
	2022
卷号	30 页码:1665-1678
通讯作者	Li, Jianxin(lijx@act.buaa.edu.cn)
摘要	. Abstractive summarization generates a concise summary to capture the key ideas of the source text. This task underpins important applications like information retrieval, document comprehension, and event tracking. While much progress has been achieved, state-of-the-art summarization approaches often fail to generate high-quality summaries to reproduce factual details accurately. One of the key limitations of existing solutions is that they are primarily concerned about extracting facts from the source text but overlook other crucial factual information, such as the related time, locations, reasons, consequences, purposes, participants and involved parties. Furthermore, the current summarization frameworks are inadequate in modeling the complex semantic relations among facts and the corresponding factual information, leaving much room for improvement. This paper presents FFSum, a novel summarization framework for exploiting multi-grained factual information to improve text summarization. To this end, FFSum constructs an individual fine-grained factual graph with multiple relations among facts and the corresponding factual information. It employs a fact-driven graph attention network to integrate multi-granular factual representations at the encoding stage. It then uses a hybrid pointer network to retrieve factual pieces from the graph for the summary generation. We evaluate the FFSum by applying it to two real-world datasets. Experimental results show that the FFSum consistently outperforms a state-of-the-art approach across evaluation datasets.
关键词	Fact consistency graph neural network language model pointer network text summarization
DOI	10.1109/TASLP.2022.3161157
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[U20B2053]
项目资助者	National Natural Science Foundation of China
WOS研究方向	Acoustics ; Engineering
WOS类目	Acoustics ; Engineering, Electrical & Electronic
WOS记录号	WOS:000795102600003
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：3[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/49450
专题	多模态人工智能系统全国重点实验室_自然语言处理
通讯作者	Li, Jianxin
作者单位	1.Beihang Univ, Beijing Adv Innovat Ctr Big Data & Brain Comp, Beijing 100190, Peoples R China 2.Beihang Univ, State Key Lab Software Dev Environm, Beijing 100190, Peoples R China 3.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100045, Peoples R China 4.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100045, Peoples R China 5.CNCERT CC, Beijing 100029, Peoples R China 6.Univ Illinois, Dept Comp Sci, Chicago, IL 60607 USA 7.Univ Leeds, Sch Comp, Leeds LS2 9JT, W Yorkshire, England
推荐引用方式 GB/T 7714	Mao, Qianren,Li, Jianxin,Peng, Hao,et al. Fact-Driven Abstractive Summarization by Utilizing Multi-Granular Multi-Relational Knowledge[J]. IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING,2022,30:1665-1678.
APA	Mao, Qianren.,Li, Jianxin.,Peng, Hao.,He, Shizhu.,Wang, Lihong.,...&Wang, Zheng.(2022).Fact-Driven Abstractive Summarization by Utilizing Multi-Granular Multi-Relational Knowledge.IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING,30,1665-1678.
MLA	Mao, Qianren,et al."Fact-Driven Abstractive Summarization by Utilizing Multi-Granular Multi-Relational Knowledge".IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING 30(2022):1665-1678.