中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 自动化研究所
  3. 中国科学院自动化研究所
  4. 多模态人工智能系统全国重点实验室
Differentiable N-gram objective on abstractive summarization

文献类型：期刊论文

作者Zhu, Yunqi2,3; Yang, Xuebing2; Wu, Yuanyuan3; Zhu, Mingjin1; Zhang, Wensheng2,3
刊名EXPERT SYSTEMS WITH APPLICATIONS
出版日期2023-04-01
卷号215页码:9
关键词Abstractive summarization Differentiable N-gram objective Neural network language model
ISSN号0957-4174
DOI10.1016/j.eswa.2022.119367
通讯作者Yang, Xuebing(yangxuebing2013@ia.ac.cn) ; Zhang, Wensheng(zhangwenshengia@hotmail.com)
英文摘要ROUGE is a standard automatic evaluation metric based on N-gram for sequence-to-sequence tasks like abstractive summarization, while cross-entropy loss is an essential objective that optimizes at unigram level for neural network language models. In this paper we present differentiable N-gram objectives, attempting to alleviate the discrepancy between training and evaluating criteria. The novelty of our work is the objective does not ceil the number of matched sub-sequences by the ground truth count of N-gram in reference sequence and weights the matched sub-sequences equally. Therefore, our proposed objective can maximize the probabilistic weight of matched sub-sequences. We jointly optimize cross-entropy loss and the objective, providing decent evaluation scores enhancement including ROUGE over abstractive summarization datasets CNN/DM and XSum, outperforming competitive N-gram objectives.
资助项目National Key R&D Program of China ; National Natural Science Foundation of China ; Hainan Provincial Natural Science Foundation of China ; [2021ZD0111000] ; [61961160707] ; [61976212] ; [622RC618]
WOS研究方向Computer Science ; Engineering ; Operations Research & Management Science
语种英语
WOS记录号WOS:000909638300001
出版者PERGAMON-ELSEVIER SCIENCE LTD
资助机构National Key R&D Program of China ; National Natural Science Foundation of China ; Hainan Provincial Natural Science Foundation of China
源URL[http://ir.ia.ac.cn/handle/173211/51123]  
专题多模态人工智能系统全国重点实验室
通讯作者Yang, Xuebing; Zhang, Wensheng
作者单位1.South China Univ Technol, Shien Ming Wu Sch Intelligent Engn, Guangzhou, Peoples R China
2.Chinese Acad Sci, Inst Automat, Beijing, Peoples R China
3.Hainan Univ, Sch Informat & Commun Engn, Haikou, Peoples R China
推荐引用方式
GB/T 7714		 Zhu, Yunqi,Yang, Xuebing,Wu, Yuanyuan,et al. Differentiable N-gram objective on abstractive summarization[J]. EXPERT SYSTEMS WITH APPLICATIONS,2023,215:9.
APA Zhu, Yunqi,Yang, Xuebing,Wu, Yuanyuan,Zhu, Mingjin,&Zhang, Wensheng.(2023).Differentiable N-gram objective on abstractive summarization.EXPERT SYSTEMS WITH APPLICATIONS,215,9.
MLA Zhu, Yunqi,et al."Differentiable N-gram objective on abstractive summarization".EXPERT SYSTEMS WITH APPLICATIONS 215(2023):9.

入库方式： OAI收割

来源：自动化研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。