CSKE: Commonsense Knowledge Enhanced Text Extension Framework for Text-Based Logical Reasoning

Yirong Zeng; Xiao Ding; Li Du; Ting Liu; Bing Qin

doi:10.1007/978-981-19-7596-7_9

CSKE: Commonsense Knowledge Enhanced Text Extension Framework for Text-Based Logical Reasoning

Yirong Zeng
, Xiao Ding^*
, Li Du
, Ting Liu
, Bing Qin

^*Corresponding author for this work

Faculty of Computing

Harbin Institute of Technology

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

Abstract

Text-based logical reasoning requires the model to understand the semantics of input text, and then understand the complex logical relationships within the text. Previous works equip pre-trained language models with the logical reasoning ability by training these models on datasets obtained by logical-driven text extension. However, these methods only generate instances based on logical expressions entailed within the input text. And we argue that external commonsense knowledge is still necessary for restoring the complete reasoning chains for generating more reasonable and abundant instances. To address this issue, in this paper, we propose CSKE, a commonsense knowledge enhanced text extension framework. CSKE incorporates abundant commonsense from an external knowledge base to restore the potentially missing logical expressions and encodes more logical relationships to then extend them through logical equivalence laws. Experiments on the benchmark datasets show that our method can improve the performance of logical reasoning, especially on the instances containing complex logical relationships.

Original language	English
Title of host publication	Knowledge Graph and Semantic Computing
Subtitle of host publication	Knowledge Graph Empowers the Digital Economy - 7th China Conference, CCKS 2022, Revised Selected Papers
Editors	Maosong Sun, Bin Xu, Guilin Qi, Kang Liu, Yubo Chen, Jiadong Ren, Yansong Feng, Yongbin Liu
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	111-122
Number of pages	12
ISBN (Print)	9789811975950
DOIs	https://doi.org/10.1007/978-981-19-7596-7_9
State	Published - 2022
Event	7th China Conference on Knowledge Graph and Semantic Computing, CCKS 2022 - Qinhuangdao, China Duration: 24 Aug 2022 → 27 Aug 2022

Publication series

Name	Communications in Computer and Information Science
Volume	1669 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	7th China Conference on Knowledge Graph and Semantic Computing, CCKS 2022
Country/Territory	China
City	Qinhuangdao
Period	24/08/22 → 27/08/22

Keywords

Commonsense knowledge
Logical expression
Logical reasoning

Access to Document

10.1007/978-981-19-7596-7_9

Cite this

Zeng, Y., Ding, X., Du, L., Liu, T., & Qin, B. (2022). CSKE: Commonsense Knowledge Enhanced Text Extension Framework for Text-Based Logical Reasoning. In M. Sun, B. Xu, G. Qi, K. Liu, Y. Chen, J. Ren, Y. Feng, & Y. Liu (Eds.), Knowledge Graph and Semantic Computing: Knowledge Graph Empowers the Digital Economy - 7th China Conference, CCKS 2022, Revised Selected Papers (pp. 111-122). (Communications in Computer and Information Science; Vol. 1669 CCIS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-19-7596-7_9

Zeng, Yirong ; Ding, Xiao ; Du, Li et al. / CSKE : Commonsense Knowledge Enhanced Text Extension Framework for Text-Based Logical Reasoning. Knowledge Graph and Semantic Computing: Knowledge Graph Empowers the Digital Economy - 7th China Conference, CCKS 2022, Revised Selected Papers. editor / Maosong Sun ; Bin Xu ; Guilin Qi ; Kang Liu ; Yubo Chen ; Jiadong Ren ; Yansong Feng ; Yongbin Liu. Springer Science and Business Media Deutschland GmbH, 2022. pp. 111-122 (Communications in Computer and Information Science).

@inproceedings{b2c636342e684e4fb48860a0eb6eab3b,

title = "CSKE: Commonsense Knowledge Enhanced Text Extension Framework for Text-Based Logical Reasoning",

abstract = "Text-based logical reasoning requires the model to understand the semantics of input text, and then understand the complex logical relationships within the text. Previous works equip pre-trained language models with the logical reasoning ability by training these models on datasets obtained by logical-driven text extension. However, these methods only generate instances based on logical expressions entailed within the input text. And we argue that external commonsense knowledge is still necessary for restoring the complete reasoning chains for generating more reasonable and abundant instances. To address this issue, in this paper, we propose CSKE, a commonsense knowledge enhanced text extension framework. CSKE incorporates abundant commonsense from an external knowledge base to restore the potentially missing logical expressions and encodes more logical relationships to then extend them through logical equivalence laws. Experiments on the benchmark datasets show that our method can improve the performance of logical reasoning, especially on the instances containing complex logical relationships.",

keywords = "Commonsense knowledge, Logical expression, Logical reasoning",

author = "Yirong Zeng and Xiao Ding and Li Du and Ting Liu and Bing Qin",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.; 7th China Conference on Knowledge Graph and Semantic Computing, CCKS 2022 ; Conference date: 24-08-2022 Through 27-08-2022",

year = "2022",

doi = "10.1007/978-981-19-7596-7\_9",

language = "英语",

isbn = "9789811975950",

series = "Communications in Computer and Information Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "111--122",

editor = "Maosong Sun and Bin Xu and Guilin Qi and Kang Liu and Yubo Chen and Jiadong Ren and Yansong Feng and Yongbin Liu",

booktitle = "Knowledge Graph and Semantic Computing",

address = "德国",

}

Zeng, Y, Ding, X, Du, L, Liu, T & Qin, B 2022, CSKE: Commonsense Knowledge Enhanced Text Extension Framework for Text-Based Logical Reasoning. in M Sun, B Xu, G Qi, K Liu, Y Chen, J Ren, Y Feng & Y Liu (eds), Knowledge Graph and Semantic Computing: Knowledge Graph Empowers the Digital Economy - 7th China Conference, CCKS 2022, Revised Selected Papers. Communications in Computer and Information Science, vol. 1669 CCIS, Springer Science and Business Media Deutschland GmbH, pp. 111-122, 7th China Conference on Knowledge Graph and Semantic Computing, CCKS 2022, Qinhuangdao, China, 24/08/22. https://doi.org/10.1007/978-981-19-7596-7_9

CSKE: Commonsense Knowledge Enhanced Text Extension Framework for Text-Based Logical Reasoning. / Zeng, Yirong; Ding, Xiao; Du, Li et al.
Knowledge Graph and Semantic Computing: Knowledge Graph Empowers the Digital Economy - 7th China Conference, CCKS 2022, Revised Selected Papers. ed. / Maosong Sun; Bin Xu; Guilin Qi; Kang Liu; Yubo Chen; Jiadong Ren; Yansong Feng; Yongbin Liu. Springer Science and Business Media Deutschland GmbH, 2022. p. 111-122 (Communications in Computer and Information Science; Vol. 1669 CCIS).

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

TY - GEN

T1 - CSKE

T2 - 7th China Conference on Knowledge Graph and Semantic Computing, CCKS 2022

AU - Zeng, Yirong

AU - Ding, Xiao

AU - Du, Li

AU - Liu, Ting

AU - Qin, Bing

PY - 2022

Y1 - 2022

N2 - Text-based logical reasoning requires the model to understand the semantics of input text, and then understand the complex logical relationships within the text. Previous works equip pre-trained language models with the logical reasoning ability by training these models on datasets obtained by logical-driven text extension. However, these methods only generate instances based on logical expressions entailed within the input text. And we argue that external commonsense knowledge is still necessary for restoring the complete reasoning chains for generating more reasonable and abundant instances. To address this issue, in this paper, we propose CSKE, a commonsense knowledge enhanced text extension framework. CSKE incorporates abundant commonsense from an external knowledge base to restore the potentially missing logical expressions and encodes more logical relationships to then extend them through logical equivalence laws. Experiments on the benchmark datasets show that our method can improve the performance of logical reasoning, especially on the instances containing complex logical relationships.

AB - Text-based logical reasoning requires the model to understand the semantics of input text, and then understand the complex logical relationships within the text. Previous works equip pre-trained language models with the logical reasoning ability by training these models on datasets obtained by logical-driven text extension. However, these methods only generate instances based on logical expressions entailed within the input text. And we argue that external commonsense knowledge is still necessary for restoring the complete reasoning chains for generating more reasonable and abundant instances. To address this issue, in this paper, we propose CSKE, a commonsense knowledge enhanced text extension framework. CSKE incorporates abundant commonsense from an external knowledge base to restore the potentially missing logical expressions and encodes more logical relationships to then extend them through logical equivalence laws. Experiments on the benchmark datasets show that our method can improve the performance of logical reasoning, especially on the instances containing complex logical relationships.

KW - Commonsense knowledge

KW - Logical expression

KW - Logical reasoning

UR - https://www.scopus.com/pages/publications/85145250621

U2 - 10.1007/978-981-19-7596-7_9

DO - 10.1007/978-981-19-7596-7_9

M3 - 会议稿件

AN - SCOPUS:85145250621

SN - 9789811975950

T3 - Communications in Computer and Information Science

SP - 111

EP - 122

BT - Knowledge Graph and Semantic Computing

A2 - Sun, Maosong

A2 - Xu, Bin

A2 - Qi, Guilin

A2 - Liu, Kang

A2 - Chen, Yubo

A2 - Ren, Jiadong

A2 - Feng, Yansong

A2 - Liu, Yongbin

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 24 August 2022 through 27 August 2022

ER -

Zeng Y, Ding X, Du L, Liu T , Qin B. CSKE: Commonsense Knowledge Enhanced Text Extension Framework for Text-Based Logical Reasoning. In Sun M, Xu B, Qi G, Liu K, Chen Y, Ren J, Feng Y, Liu Y, editors, Knowledge Graph and Semantic Computing: Knowledge Graph Empowers the Digital Economy - 7th China Conference, CCKS 2022, Revised Selected Papers. Springer Science and Business Media Deutschland GmbH. 2022. p. 111-122. (Communications in Computer and Information Science). doi: 10.1007/978-981-19-7596-7_9

CSKE: Commonsense Knowledge Enhanced Text Extension Framework for Text-Based Logical Reasoning

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this