| 90 | 1 | 112 |
| Downloads | Citas | Reads |
Most machine reading comprehension(MRC) models were based on end-to-end deep learning networks with various attention-mechanisms, but such models would lose sentence-level semantic information. Additionally, complex reasoning was unnecessary for answering questions in existing datasets, and the answers were only related to a few sentences in the background passages. Based on this, machine reading comprehension models were proposed to be divided into two stages. The first stage searched sentences related to questions and generated new background passages. The second stage then extracted answers based on these reduced passages. Experiments results confirmed that the prediction performance was improved after locating the related sentences. The SQuAD MRC dataset was also divided into two parts to adapt to the training of the new framework. The new datasets were also used to test the influence of the scale of the relevant content on the performance of MRC models.
[1] 张超然，裘杭萍，孙毅，等.基于预训练模型的机器阅读理解研究综述[J].计算机工程与应用，2020,56(11):17-25.ZHANG C R,QIU H P,SUN Y,et al.Review of machine reading comprehension based on pre-training language model[J].Computer engineering and applications,2020,56(11):17-25.
[2] RICHARDSON M,BURGES C,RENSHAW E.MCTest:a challenge dataset for the open-domain machine comprehension of text[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing.Stroudsburg:Association for Computational Linguistics,2013:193-203.
[3] CHEN D Q,BOLTON J,MANNING C D.A thorough examination of the CNN/Daily Mail reading comprehension task[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics.Stroudsburg:Association for Computational Linguistics,2016:2358-2367.
[4] LAI G K,XIE Q Z,LIU H X,et al.RACE:large-scale reading comprehension dataset from examinations[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing.Stroudsburg:Association for Computational Linguistics,2017:785-794.
[5] RAJPURKAR P,ZHANG J,LOPYREV K,et al.SQuAD:100 000+ questions for machine comprehension of text[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing.Stroudsburg:Association for Computational Linguistics,2016:2383-2392.
[6] HE W,LIU K,LIU J,et al.DuReader:a Chinese machine reading comprehension dataset from real-world applications[C]//Proceedings of the Workshop on Machine Reading for Question Answering.Stroudsburg:Association for Computational Linguistics,2018:37-46.
[7] MIKOLOV T,CHEN K,CORRADO G,et al.Efficient estimation of word representations in vector space[EB/OL].[2020-04-25].https://arxiv.org/pdf/1301.3781.pdf.
[8] PETERS M E,NEUMANN M,IYYER M,et al.Deep contextualized word representations[C]//Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics.Stroudsburg:Association for Computational Linguistics,2018:2227-2237.
[9] MCCANN B,BRADBURY J,XIONG C,et al.Learned in translation:contextualized word vectors[C]//Proceedings of the Advances in Neural Information Processing Systems.Cambridge:MIT Press,2017:6294-6305.
[10] VASWANI A,SHAZEER N,PARMAR N,et al.Attention is all you need[C]//Proceedings of the Advances in Neural Information Processing Systems.Cambridge:MIT Press,2017:5998-6008.
[11] MINJOON S,ANIRUDDHA K,ALI F,et al.Bidirectional attention flow for machine comprehension[EB/OL].[2020-05-06].https://arxiv.org/pdf/1611.01603.pdf.
[12] ADAMS W Y,DAVID D,MINH T L,et al.QANet:combining local convolution with global self-attention for reading comprehension[EB/OL].[2020-05-06].https://arxiv.org/pdf/1804.09541.pdf.
[13] YANG Z,DAI Z,YANG Y,et al.XLNet:generalized autoregressive pretraining for language understanding[C]//Proceedings of the Advances in Neural Information Processing Systems.Cambridge:MIT Press,2019:5753-5763.
[14] ZHANG Z Y,HAN X,LIU Z Y,et al.ERNIE:enhanced language representation with informative entities[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics.Stroudsburg:Association for Computational Linguistics,2019:1441-1451.
Basic Information:
DOI：10.13705/j.issn.1671-6841.2020316
China Classification Code:TP391.1
Citation Information:
[1]DENG Chaoyu,ZHAO Shan,XIAO Xiaoqiang ,et al.A Method of Machine Reading Comprehension Based on Explicit Positioning[J].Journal of Zhengzhou University(Natural Science Edition),2021,53(03):37-41+49.DOI:10.13705/j.issn.1671-6841.2020316.
Fund Information:
国家重点研发计划项目(2020YFC2003400,SQ2019ZD090149); 国家自然科学基金项目(62072465); 国家科技重大专项重大新药创制(2018ZX09201-014)
2020-10-04
2020
2020-11-18
2021-07-13
2021
1
2020-12-24
2020-12-24
2020-12-24