Machine reading comprehension: Matching and orders

Ao Liu; Lizhen Qu; Junyu Lu; Chenbin Zhang; Zenglin Xu

doi:10.1145/3357384.3358139

Machine reading comprehension: Matching and orders

Ao Liu, Lizhen Qu, Junyu Lu, Chenbin Zhang, Zenglin Xu

Department of Data Science & AI

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

5 Citations (Scopus)

Abstract

In this paper, we study the machine reading comprehension of temporal order in text. Given a document of instruction sequences, a model aims to find out the most coherent sequences of activities matching the document among all answer candidates. To tackle the task, we propose OrdMatch model, which is able to match each activity in a sequence to the corresponding instruction in the document and regularizes the partial order of activities to match the order of instructions. We evaluate the task using the RecipeQA dataset, which includes step-by-step instructions of cooking recipes. Our model outperforms the state-of-the-art models with a wide margin. The experimental results demonstrate the effectiveness of our novel ordering regularizer. Our code will be made available at https://github.com/Aolius/OrdMatch.

Original language	English
Title of host publication	Proceedings of the 28th ACM International Conference on Information and Knowledge Management
Editors	Meng Jiang, Mu Qiao
Place of Publication	New York NY USA
Publisher	Association for Computing Machinery (ACM)
Pages	2057-2060
Number of pages	4
ISBN (Electronic)	9781450369763
DOIs	https://doi.org/10.1145/3357384.3358139
Publication status	Published - 2019
Event	ACM International Conference on Information and Knowledge Management 2019 - Beijing, China Duration: 3 Nov 2019 → 7 Nov 2019 Conference number: 28th http://www.cikm2019.net/ https://dl.acm.org/doi/proceedings/10.1145/3357384

Conference

Conference	ACM International Conference on Information and Knowledge Management 2019
Abbreviated title	CIKM 2019
Country/Territory	China
City	Beijing
Period	3/11/19 → 7/11/19
Internet address	http://www.cikm2019.net/ https://dl.acm.org/doi/proceedings/10.1145/3357384

Keywords

Activity Ordering
Machine Reading Comprehension
RecipeQA

Access to Document

10.1145/3357384.3358139

Cite this

@inproceedings{c63e2d0e3c3c4343a7982d4098a0a2b3,

title = "Machine reading comprehension: Matching and orders",

abstract = "In this paper, we study the machine reading comprehension of temporal order in text. Given a document of instruction sequences, a model aims to find out the most coherent sequences of activities matching the document among all answer candidates. To tackle the task, we propose OrdMatch model, which is able to match each activity in a sequence to the corresponding instruction in the document and regularizes the partial order of activities to match the order of instructions. We evaluate the task using the RecipeQA dataset, which includes step-by-step instructions of cooking recipes. Our model outperforms the state-of-the-art models with a wide margin. The experimental results demonstrate the effectiveness of our novel ordering regularizer. Our code will be made available at https://github.com/Aolius/OrdMatch.",

keywords = "Activity Ordering, Machine Reading Comprehension, RecipeQA",

author = "Ao Liu and Lizhen Qu and Junyu Lu and Chenbin Zhang and Zenglin Xu",

note = "Funding Information: This work was partially supported by NSF China (Project No. 615721 11) and a Research Fund for the Central Universities of China (Project No. ZYGX2016Z003). Publisher Copyright: {\textcopyright} 2019 Association for Computing Machinery.; ACM International Conference on Information and Knowledge Management 2019, CIKM 2019 ; Conference date: 03-11-2019 Through 07-11-2019",

year = "2019",

doi = "10.1145/3357384.3358139",

language = "English",

pages = "2057--2060",

editor = "Meng Jiang and Mu Qiao",

booktitle = "Proceedings of the 28th ACM International Conference on Information and Knowledge Management",

publisher = "Association for Computing Machinery (ACM)",

address = "United States of America",

url = "http://www.cikm2019.net/, https://dl.acm.org/doi/proceedings/10.1145/3357384",

}

Liu, A, Qu, L, Lu, J, Zhang, C & Xu, Z 2019, Machine reading comprehension: Matching and orders. in M Jiang & M Qiao (eds), Proceedings of the 28th ACM International Conference on Information and Knowledge Management. Association for Computing Machinery (ACM), New York NY USA, pp. 2057-2060, ACM International Conference on Information and Knowledge Management 2019, Beijing, China, 3/11/19. https://doi.org/10.1145/3357384.3358139

Machine reading comprehension: Matching and orders. / Liu, Ao; Qu, Lizhen; Lu, Junyu et al.
Proceedings of the 28th ACM International Conference on Information and Knowledge Management. ed. / Meng Jiang; Mu Qiao. New York NY USA: Association for Computing Machinery (ACM), 2019. p. 2057-2060.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Machine reading comprehension

T2 - ACM International Conference on Information and Knowledge Management 2019

AU - Liu, Ao

AU - Qu, Lizhen

AU - Lu, Junyu

AU - Zhang, Chenbin

AU - Xu, Zenglin

N1 - Conference code: 28th

PY - 2019

Y1 - 2019

N2 - In this paper, we study the machine reading comprehension of temporal order in text. Given a document of instruction sequences, a model aims to find out the most coherent sequences of activities matching the document among all answer candidates. To tackle the task, we propose OrdMatch model, which is able to match each activity in a sequence to the corresponding instruction in the document and regularizes the partial order of activities to match the order of instructions. We evaluate the task using the RecipeQA dataset, which includes step-by-step instructions of cooking recipes. Our model outperforms the state-of-the-art models with a wide margin. The experimental results demonstrate the effectiveness of our novel ordering regularizer. Our code will be made available at https://github.com/Aolius/OrdMatch.

AB - In this paper, we study the machine reading comprehension of temporal order in text. Given a document of instruction sequences, a model aims to find out the most coherent sequences of activities matching the document among all answer candidates. To tackle the task, we propose OrdMatch model, which is able to match each activity in a sequence to the corresponding instruction in the document and regularizes the partial order of activities to match the order of instructions. We evaluate the task using the RecipeQA dataset, which includes step-by-step instructions of cooking recipes. Our model outperforms the state-of-the-art models with a wide margin. The experimental results demonstrate the effectiveness of our novel ordering regularizer. Our code will be made available at https://github.com/Aolius/OrdMatch.

KW - Activity Ordering

KW - Machine Reading Comprehension

KW - RecipeQA

UR - http://www.scopus.com/inward/record.url?scp=85075443852&partnerID=8YFLogxK

U2 - 10.1145/3357384.3358139

DO - 10.1145/3357384.3358139

M3 - Conference Paper

AN - SCOPUS:85075443852

SP - 2057

EP - 2060

BT - Proceedings of the 28th ACM International Conference on Information and Knowledge Management

A2 - Jiang, Meng

A2 - Qiao, Mu

PB - Association for Computing Machinery (ACM)

CY - New York NY USA

Y2 - 3 November 2019 through 7 November 2019

ER -

Machine reading comprehension: Matching and orders

Abstract

Conference

Keywords

Access to Document

Other files and links

Cite this