Compositional Generalization in Multilingual Semantic Parsing over Wikidata

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

Compositional Generalization in Multilingual Semantic Parsing over Wikidata. / Cui, Ruixiang; Aralikatte, Rahul; Lent, Heather; Hershcovich, Daniel.

In: Transactions of the Association for Computational Linguistics, Vol. 10, 2022, p. 937-955.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Cui, R, Aralikatte, R, Lent, H & Hershcovich, D 2022, 'Compositional Generalization in Multilingual Semantic Parsing over Wikidata', Transactions of the Association for Computational Linguistics, vol. 10, pp. 937-955. https://doi.org/10.1162/tacl_a_00499

APA

Cui, R., Aralikatte, R., Lent, H., & Hershcovich, D. (2022). Compositional Generalization in Multilingual Semantic Parsing over Wikidata. Transactions of the Association for Computational Linguistics, 10, 937-955. https://doi.org/10.1162/tacl_a_00499

Vancouver

Cui R, Aralikatte R, Lent H, Hershcovich D. Compositional Generalization in Multilingual Semantic Parsing over Wikidata. Transactions of the Association for Computational Linguistics. 2022;10:937-955. https://doi.org/10.1162/tacl_a_00499

Author

Cui, Ruixiang ; Aralikatte, Rahul ; Lent, Heather ; Hershcovich, Daniel. / Compositional Generalization in Multilingual Semantic Parsing over Wikidata. In: Transactions of the Association for Computational Linguistics. 2022 ; Vol. 10. pp. 937-955.

Bibtex

@article{0566ccc6979a4e0b97cc74bb982f1bb5,
title = "Compositional Generalization in Multilingual Semantic Parsing over Wikidata",
abstract = "Semantic parsing (SP) allows humans to lever-age vast knowledge resources through natural interaction. However, parsers are mostly de-signed for and evaluated on English resources, such as CFQ (Keysers et al., 2020), the current standard benchmark based on English data generated from grammar rules and oriented towards Freebase, an outdated knowledge base. We propose a method for creating a multilingual, parallel dataset of question-query pairs, grounded in Wikidata. We introduce such a dataset, which we call Multilingual Compositional Wikidata Questions (MCWQ), and use it to analyze the compositional generalization of semantic parsers in Hebrew, Kannada, Chinese, and English. While within-language generalization is comparable across languages, experiments on zero-shot cross-lingual transfer demonstrate that cross-lingual compositional generalization fails, even with state-of-the-art pretrained multilingual encod-ers. Furthermore, our methodology, dataset, and results will facilitate future research on SP in more realistic and diverse settings than has been possible with existing resources.",
author = "Ruixiang Cui and Rahul Aralikatte and Heather Lent and Daniel Hershcovich",
note = "Publisher Copyright: {\textcopyright} 2022 Association for Computational Linguistics.",
year = "2022",
doi = "10.1162/tacl_a_00499",
language = "English",
volume = "10",
pages = "937--955",
journal = "Transactions of the Association for Computational Linguistics",
issn = "2307-387X",
publisher = "MIT Press",

}

RIS

TY - JOUR

T1 - Compositional Generalization in Multilingual Semantic Parsing over Wikidata

AU - Cui, Ruixiang

AU - Aralikatte, Rahul

AU - Lent, Heather

AU - Hershcovich, Daniel

N1 - Publisher Copyright: © 2022 Association for Computational Linguistics.

PY - 2022

Y1 - 2022

N2 - Semantic parsing (SP) allows humans to lever-age vast knowledge resources through natural interaction. However, parsers are mostly de-signed for and evaluated on English resources, such as CFQ (Keysers et al., 2020), the current standard benchmark based on English data generated from grammar rules and oriented towards Freebase, an outdated knowledge base. We propose a method for creating a multilingual, parallel dataset of question-query pairs, grounded in Wikidata. We introduce such a dataset, which we call Multilingual Compositional Wikidata Questions (MCWQ), and use it to analyze the compositional generalization of semantic parsers in Hebrew, Kannada, Chinese, and English. While within-language generalization is comparable across languages, experiments on zero-shot cross-lingual transfer demonstrate that cross-lingual compositional generalization fails, even with state-of-the-art pretrained multilingual encod-ers. Furthermore, our methodology, dataset, and results will facilitate future research on SP in more realistic and diverse settings than has been possible with existing resources.

AB - Semantic parsing (SP) allows humans to lever-age vast knowledge resources through natural interaction. However, parsers are mostly de-signed for and evaluated on English resources, such as CFQ (Keysers et al., 2020), the current standard benchmark based on English data generated from grammar rules and oriented towards Freebase, an outdated knowledge base. We propose a method for creating a multilingual, parallel dataset of question-query pairs, grounded in Wikidata. We introduce such a dataset, which we call Multilingual Compositional Wikidata Questions (MCWQ), and use it to analyze the compositional generalization of semantic parsers in Hebrew, Kannada, Chinese, and English. While within-language generalization is comparable across languages, experiments on zero-shot cross-lingual transfer demonstrate that cross-lingual compositional generalization fails, even with state-of-the-art pretrained multilingual encod-ers. Furthermore, our methodology, dataset, and results will facilitate future research on SP in more realistic and diverse settings than has been possible with existing resources.

UR - http://www.scopus.com/inward/record.url?scp=85138287503&partnerID=8YFLogxK

U2 - 10.1162/tacl_a_00499

DO - 10.1162/tacl_a_00499

M3 - Journal article

AN - SCOPUS:85138287503

VL - 10

SP - 937

EP - 955

JO - Transactions of the Association for Computational Linguistics

JF - Transactions of the Association for Computational Linguistics

SN - 2307-387X

ER -

ID: 322574313