Reanalyzing L2 Preposition Learning with Bayesian Mixed Effects and a Pretrained Language Model

Jakob Prange; Man Ho Ivy Wong

doi:10.18653/v1/2023.acl-long.712

Reanalyzing L2 Preposition Learning with Bayesian Mixed Effects and a Pretrained Language Model

Jakob Prange, Man Ho Ivy Wong

Department of Chinese and Bilingual Studies

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

1 Citation (Scopus)

Abstract

We use both Bayesian and neural models to dissect a data set of Chinese learners’ pre- and post-interventional responses to two tests measuring their understanding of English prepositions. The results mostly replicate previous findings from frequentist analyses and newly reveal crucial interactions between student ability, task type, and stimulus sentence. Given the sparsity of the data as well as high diversity among learners, the Bayesian method proves most useful; but we also see potential in using language model probabilities as predictors of grammaticality and learnability.

Original language	English
Title of host publication	Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Editors	Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Publisher	Association for Computational Linguistics (ACL)
Pages	12722-12736
ISBN (Electronic)	978-1-959429-72-2
DOIs	https://doi.org/10.18653/v1/2023.acl-long.712
Publication status	Published - Jul 2023
Event	61st Annual Meeting of the Association for Computational Linguistics, ACL 2023 - Toronto, Canada Duration: 9 Jul 2023 → 14 Jul 2023

Conference

Conference	61st Annual Meeting of the Association for Computational Linguistics, ACL 2023
Country/Territory	Canada
City	Toronto
Period	9/07/23 → 14/07/23

More information

10.18653/v1/2023.acl-long.712

Cite this

Prange, J., & Wong, M. H. I. (2023). Reanalyzing L2 Preposition Learning with Bayesian Mixed Effects and a Pretrained Language Model. In A. Rogers, J. Boyd-Graber, & N. Okazaki (Eds.), Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 12722-12736). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/2023.acl-long.712

Prange, Jakob ; Wong, Man Ho Ivy. / Reanalyzing L2 Preposition Learning with Bayesian Mixed Effects and a Pretrained Language Model. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). editor / Anna Rogers ; Jordan Boyd-Graber ; Naoaki Okazaki. Association for Computational Linguistics (ACL), 2023. pp. 12722-12736

@inproceedings{c6ee8e12a99040d8855c46b2321f092f,

title = "Reanalyzing L2 Preposition Learning with Bayesian Mixed Effects and a Pretrained Language Model",

abstract = "We use both Bayesian and neural models to dissect a data set of Chinese learners{\textquoteright} pre- and post-interventional responses to two tests measuring their understanding of English prepositions. The results mostly replicate previous findings from frequentist analyses and newly reveal crucial interactions between student ability, task type, and stimulus sentence. Given the sparsity of the data as well as high diversity among learners, the Bayesian method proves most useful; but we also see potential in using language model probabilities as predictors of grammaticality and learnability.",

author = "Jakob Prange and Wong, {Man Ho Ivy}",

year = "2023",

month = jul,

doi = "10.18653/v1/2023.acl-long.712",

language = "English",

pages = "12722--12736",

editor = "Anna Rogers and Jordan Boyd-Graber and Naoaki Okazaki",

booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",

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

note = "61st Annual Meeting of the Association for Computational Linguistics, ACL 2023 ; Conference date: 09-07-2023 Through 14-07-2023",

}

Prange, J & Wong, MHI 2023, Reanalyzing L2 Preposition Learning with Bayesian Mixed Effects and a Pretrained Language Model. in A Rogers, J Boyd-Graber & N Okazaki (eds), Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Association for Computational Linguistics (ACL), pp. 12722-12736, 61st Annual Meeting of the Association for Computational Linguistics, ACL 2023, Toronto, Canada, 9/07/23. https://doi.org/10.18653/v1/2023.acl-long.712

Reanalyzing L2 Preposition Learning with Bayesian Mixed Effects and a Pretrained Language Model. / Prange, Jakob; Wong, Man Ho Ivy.
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). ed. / Anna Rogers; Jordan Boyd-Graber; Naoaki Okazaki. Association for Computational Linguistics (ACL), 2023. p. 12722-12736.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Reanalyzing L2 Preposition Learning with Bayesian Mixed Effects and a Pretrained Language Model

AU - Prange, Jakob

AU - Wong, Man Ho Ivy

PY - 2023/7

Y1 - 2023/7

N2 - We use both Bayesian and neural models to dissect a data set of Chinese learners’ pre- and post-interventional responses to two tests measuring their understanding of English prepositions. The results mostly replicate previous findings from frequentist analyses and newly reveal crucial interactions between student ability, task type, and stimulus sentence. Given the sparsity of the data as well as high diversity among learners, the Bayesian method proves most useful; but we also see potential in using language model probabilities as predictors of grammaticality and learnability.

AB - We use both Bayesian and neural models to dissect a data set of Chinese learners’ pre- and post-interventional responses to two tests measuring their understanding of English prepositions. The results mostly replicate previous findings from frequentist analyses and newly reveal crucial interactions between student ability, task type, and stimulus sentence. Given the sparsity of the data as well as high diversity among learners, the Bayesian method proves most useful; but we also see potential in using language model probabilities as predictors of grammaticality and learnability.

U2 - 10.18653/v1/2023.acl-long.712

DO - 10.18653/v1/2023.acl-long.712

M3 - Conference article published in proceeding or book

SP - 12722

EP - 12736

BT - Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

A2 - Rogers, Anna

A2 - Boyd-Graber, Jordan

A2 - Okazaki, Naoaki

PB - Association for Computational Linguistics (ACL)

T2 - 61st Annual Meeting of the Association for Computational Linguistics, ACL 2023

Y2 - 9 July 2023 through 14 July 2023

ER -

Prange J, Wong MHI. Reanalyzing L2 Preposition Learning with Bayesian Mixed Effects and a Pretrained Language Model. In Rogers A, Boyd-Graber J, Okazaki N, editors, Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Association for Computational Linguistics (ACL). 2023. p. 12722-12736 doi: 10.18653/v1/2023.acl-long.712

Reanalyzing L2 Preposition Learning with Bayesian Mixed Effects and a Pretrained Language Model

Abstract

Conference

More information

Fingerprint

Cite this