Document Modeling with Graph Attention Networks for Multi-grained Machine Reading Comprehension

Bo Zheng, Haoyang Wen, Yaobo Liang, Nan Duan, Wanxiang Che, Daxin Jiang, Ming Zhou, Ting Liu


Abstract
Natural Questions is a new challenging machine reading comprehension benchmark with two-grained answers, which are a long answer (typically a paragraph) and a short answer (one or more entities inside the long answer). Despite the effectiveness of existing methods on this benchmark, they treat these two sub-tasks individually during training while ignoring their dependencies. To address this issue, we present a novel multi-grained machine reading comprehension framework that focuses on modeling documents at their hierarchical nature, which are different levels of granularity: documents, paragraphs, sentences, and tokens. We utilize graph attention networks to obtain different levels of representations so that they can be learned simultaneously. The long and short answers can be extracted from paragraph-level representation and token-level representation, respectively. In this way, we can model the dependencies between the two-grained answers to provide evidence for each other. We jointly train the two sub-tasks, and our experiments show that our approach significantly outperforms previous systems at both long and short answer criteria.
Anthology ID:
2020.acl-main.599
Volume:
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2020
Address:
Online
Editors:
Dan Jurafsky, Joyce Chai, Natalie Schluter, Joel Tetreault
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6708–6718
Language:
URL:
https://aclanthology.org/2020.acl-main.599
DOI:
10.18653/v1/2020.acl-main.599
Bibkey:
Cite (ACL):
Bo Zheng, Haoyang Wen, Yaobo Liang, Nan Duan, Wanxiang Che, Daxin Jiang, Ming Zhou, and Ting Liu. 2020. Document Modeling with Graph Attention Networks for Multi-grained Machine Reading Comprehension. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 6708–6718, Online. Association for Computational Linguistics.
Cite (Informal):
Document Modeling with Graph Attention Networks for Multi-grained Machine Reading Comprehension (Zheng et al., ACL 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/dois-2013-emnlp/2020.acl-main.599.pdf
Video:
 http://slideslive.com/38928861
Code
 DancingSoul/NQ_BERT-DM
Data
SQuAD