Table to text generation with accurate content copying

被引:0
|
作者
Yang Yang
Juan Cao
Yujun Wen
Pengzhou Zhang
机构
[1] Communication University of China,State Key Laboratory of Media Convergence and Communication
来源
Scientific Reports | / 11卷
关键词
D O I
暂无
中图分类号
学科分类号
摘要
Generating fluent, coherent, and informative text from structured data is called table-to-text generation. Copying words from the table is a common method to solve the “out-of-vocabulary” problem, but it’s difficult to achieve accurate copying. In order to overcome this problem, we invent an auto-regressive framework based on the transformer that combines a copying mechanism and language modeling to generate target texts. Firstly, to make the model better learn the semantic relevance between table and text, we apply a word transformation method, which incorporates the field and position information into the target text to acquire the position of where to copy. Then we propose two auxiliary learning objectives, namely table-text constraint loss and copy loss. Table-text constraint loss is used to effectively model table inputs, whereas copy loss is exploited to precisely copy word fragments from a table. Furthermore, we improve the text search strategy to reduce the probability of generating incoherent and repetitive sentences. The model is verified by experiments on two datasets and better results are obtained than the baseline model. On WIKIBIO, the result is improved from 45.47 to 46.87 on BLEU and from 41.54 to 42.28 on ROUGE. On ROTOWIRE, the result is increased by 4.29% on CO metric, and 1.93 points higher on BLEU.
引用
收藏
相关论文
共 50 条
  • [41] A fast algorithm for accurate content-adaptive mesh generation
    Yang, YY
    Wernick, MN
    Brankov, JG
    2001 INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, VOL III, PROCEEDINGS, 2001, : 868 - 871
  • [42] A fast approach for accurate content-adaptive mesh generation
    Yang, YY
    Wernick, MN
    Brankov, JG
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2003, 12 (08) : 866 - 881
  • [43] Select and Attend: Towards Controllable Content Selection in Text Generation
    Shen, Xiaoyu
    Suzuki, Jun
    Inui, Kentaro
    Su, Hui
    Klakow, Dietrich
    Sekine, Satoshi
    2019 CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING AND THE 9TH INTERNATIONAL JOINT CONFERENCE ON NATURAL LANGUAGE PROCESSING (EMNLP-IJCNLP 2019): PROCEEDINGS OF THE CONFERENCE, 2019, : 579 - 590
  • [44] Neural data-to-text generation with dynamic content planning
    Chen, Kai
    Li, Fayuan
    Hu, Baotian
    Peng, Weihua
    Chen, Qingcai
    Yu, Hong
    Xiang, Yang
    KNOWLEDGE-BASED SYSTEMS, 2021, 215
  • [45] Elaborative Simplification: Content Addition and Explanation Generation in Text Simplification
    Srikanth, Neha
    Li, Junyi Jessy
    FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, ACL-IJCNLP 2021, 2021, : 5123 - 5137
  • [46] Efficiency of automatic text generators for online review content generation
    Perez-Castro, A.
    Martinez-Torres, M. R.
    Toral, S. L.
    TECHNOLOGICAL FORECASTING AND SOCIAL CHANGE, 2023, 189
  • [47] A Data-to-Text Generation Model with Deduplicated Content Planning
    Wang, Mengda
    Cao, Jianjun
    Yu, Xu
    Nie, Zibo
    BIG DATA, BIGDATA 2022, 2022, 1709 : 92 - 103
  • [48] Towards Accurate Text-based Image Captioning with Content Diversity Exploration
    Xu, Guanghui
    Niu, Shuaicheng
    Tan, Mingkui
    Luo, Yucheng
    Du, Qing
    Wu, Qi
    2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, 2021, : 12632 - 12641
  • [49] Joint Copying and Restricted Generation for Paraphrase
    Cao, Ziqiang
    Luo, Chuwei
    Li, Wenjie
    Li, Sujian
    THIRTY-FIRST AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2017, : 3152 - 3158
  • [50] COPYING AND COMPOSING - TEXT AND CONTEXT IN CHILDRENS INFORMATIONAL WRITING
    COLLERSON, J
    EDUCATIONAL REVIEW, 1986, 38 (02) : 139 - 150