Cross-lingual Inference with a Chinese Entailment Graph

Tianyi Li; Sabine Weber; Javad Hosseini; Liane Guillou; Mark Steedman

doi:10.18653/v1/2022.findings-acl.96

Cross-lingual Inference with a Chinese Entailment Graph

Tianyi Li, Sabine Weber, Javad Hosseini, Liane Guillou, Mark Steedman

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

Abstract / Description of output

Predicate entailment detection is a crucial task for question-answering from text, where previous work has explored unsupervised learning of entailment graphs from typed open relation triples. In this paper, we present the first pipeline for building Chinese entailment graphs, which involves a novel high-recall open relation extraction (ORE) method and the first Chinese fine-grained entity typing dataset under the FIGER type ontology. Through experiments on the Levy-Holt dataset, we verify the strength of our Chinese entailment graph, and reveal the cross-lingual complementarity: on the parallel Levy-Holt dataset, an ensemble of Chinese and English entailment graphs outperforms both monolingual graphs, and raises unsupervised SOTA by 4.7 AUC points.

Original language	English
Title of host publication	Findings of the Association for Computational Linguistics: ACL 2022
Publisher	Association for Computational Linguistics (ACL)
Pages	1214-1233
Number of pages	20
ISBN (Print)	9781955917254
DOIs	https://doi.org/10.18653/v1/2022.findings-acl.96
Publication status	Published - 22 May 2022
Event	60th Annual Meeting of the Association for Computational Linguistics - The Convention Centre Dublin, Dublin, Ireland Duration: 22 May 2022 → 27 May 2022 https://www.2022.aclweb.org

Conference

Conference	60th Annual Meeting of the Association for Computational Linguistics
Abbreviated title	ACL 2022
Country/Territory	Ireland
City	Dublin
Period	22/05/22 → 27/05/22
Internet address	https://www.2022.aclweb.org

Access to Document

10.18653/v1/2022.findings-acl.96Licence: Creative Commons: Attribution (CC-BY)

Cross-lingual_LI_DOA23022022_VOR_CC BYFinal published version, 1.04 MBLicence: Creative Commons: Attribution (CC-BY)

https://aclanthology.org/2022.findings-acl.96/Licence: Creative Commons: Attribution (CC-BY)

SEMANTAX-Form-Independent Semantics for Natural Language Understanding
Steedman, M.
EU government bodies
1/08/17 → 31/07/23
Project: Research

File

Cite this

@inproceedings{89ff962f76884d5f9533c28fcd84d454,

title = "Cross-lingual Inference with a Chinese Entailment Graph",

abstract = "Predicate entailment detection is a crucial task for question-answering from text, where previous work has explored unsupervised learning of entailment graphs from typed open relation triples. In this paper, we present the first pipeline for building Chinese entailment graphs, which involves a novel high-recall open relation extraction (ORE) method and the first Chinese fine-grained entity typing dataset under the FIGER type ontology. Through experiments on the Levy-Holt dataset, we verify the strength of our Chinese entailment graph, and reveal the cross-lingual complementarity: on the parallel Levy-Holt dataset, an ensemble of Chinese and English entailment graphs outperforms both monolingual graphs, and raises unsupervised SOTA by 4.7 AUC points.",

author = "Tianyi Li and Sabine Weber and Javad Hosseini and Liane Guillou and Mark Steedman",

note = "This work was supported partly by ERC Advanced Fellowship GA 742137 SEMANTAX, a Mozilla PhD scholarship at Informatics Graduate School and the University of Edinburgh Huawei Laboratory.; 60th Annual Meeting of the Association for Computational Linguistics, ACL 2022 ; Conference date: 22-05-2022 Through 27-05-2022",

year = "2022",

month = may,

day = "22",

doi = "10.18653/v1/2022.findings-acl.96",

language = "English",

isbn = "9781955917254",

pages = "1214--1233",

booktitle = "Findings of the Association for Computational Linguistics: ACL 2022",

publisher = "Association for Computational Linguistics (ACL)",

url = "https://www.2022.aclweb.org",

}

Li, T, Weber, S, Hosseini, J, Guillou, L & Steedman, M 2022, Cross-lingual Inference with a Chinese Entailment Graph. in Findings of the Association for Computational Linguistics: ACL 2022. Association for Computational Linguistics (ACL), pp. 1214-1233, 60th Annual Meeting of the Association for Computational Linguistics, Dublin, Ireland, 22/05/22. https://doi.org/10.18653/v1/2022.findings-acl.96

TY - GEN

T1 - Cross-lingual Inference with a Chinese Entailment Graph

AU - Li, Tianyi

AU - Weber, Sabine

AU - Hosseini, Javad

AU - Guillou, Liane

AU - Steedman, Mark

N1 - This work was supported partly by ERC Advanced Fellowship GA 742137 SEMANTAX, a Mozilla PhD scholarship at Informatics Graduate School and the University of Edinburgh Huawei Laboratory.

PY - 2022/5/22

Y1 - 2022/5/22

N2 - Predicate entailment detection is a crucial task for question-answering from text, where previous work has explored unsupervised learning of entailment graphs from typed open relation triples. In this paper, we present the first pipeline for building Chinese entailment graphs, which involves a novel high-recall open relation extraction (ORE) method and the first Chinese fine-grained entity typing dataset under the FIGER type ontology. Through experiments on the Levy-Holt dataset, we verify the strength of our Chinese entailment graph, and reveal the cross-lingual complementarity: on the parallel Levy-Holt dataset, an ensemble of Chinese and English entailment graphs outperforms both monolingual graphs, and raises unsupervised SOTA by 4.7 AUC points.

AB - Predicate entailment detection is a crucial task for question-answering from text, where previous work has explored unsupervised learning of entailment graphs from typed open relation triples. In this paper, we present the first pipeline for building Chinese entailment graphs, which involves a novel high-recall open relation extraction (ORE) method and the first Chinese fine-grained entity typing dataset under the FIGER type ontology. Through experiments on the Levy-Holt dataset, we verify the strength of our Chinese entailment graph, and reveal the cross-lingual complementarity: on the parallel Levy-Holt dataset, an ensemble of Chinese and English entailment graphs outperforms both monolingual graphs, and raises unsupervised SOTA by 4.7 AUC points.

U2 - 10.18653/v1/2022.findings-acl.96

DO - 10.18653/v1/2022.findings-acl.96

M3 - Conference contribution

SN - 9781955917254

SP - 1214

EP - 1233

BT - Findings of the Association for Computational Linguistics: ACL 2022

PB - Association for Computational Linguistics (ACL)

T2 - 60th Annual Meeting of the Association for Computational Linguistics

Y2 - 22 May 2022 through 27 May 2022

ER -

Cross-lingual Inference with a Chinese Entailment Graph

Abstract / Description of output

Conference

Access to Document

Fingerprint

Projects

SEMANTAX-Form-Independent Semantics for Natural Language Understanding

Cite this