Exploring Nominal Coercion in Semantic Spaces with Static and Contextualized Word Embeddings

Chenxin Liu; Emmanuele Chersoni

Exploring Nominal Coercion in Semantic Spaces with Static and Contextualized Word Embeddings

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

Abstract

The distinction between mass nouns and count nouns has a long history in formal semantics, and linguists have been trying to identify the semantic properties defining the two classes. However, they also recognized that both can undergo meaning shifts and be used in contexts of a different type, via nominal coercion. In this paper, we present an approach to measure the meaning shift in count-mass coercion in English that makes use of static and contextualized word embedding distance. Our results show that the coercion shifts are detected only by a small subset of the traditional word embedding models, and that the shifts detected by the contextualized embedding of BERT are more pronounced for mass nouns.

Original language	English
Title of host publication	Proceedings of the Workshop on Cognitive Aspects of the Lexicon (CogALex-VII)
Editors	Michael Zock, Emmanuele Chersoni, Yu-Yin Hsu, Enrico Santus
Publisher	Association for Computational Linguistics, ACL Anthology
Pages	49–57
Number of pages	9
ISBN (Print)	978-1-959429-01-2
Publication status	Published - Nov 2022
Event	AACL-IJCNLP Workshop on Cognitive Aspects of the Lexicon 2022 - Online Duration: 20 Nov 2022 → 20 Nov 2022

Conference

Conference	AACL-IJCNLP Workshop on Cognitive Aspects of the Lexicon 2022
Abbreviated title	CogALex 2022
Period	20/11/22 → 20/11/22

More information

https://aclanthology.org/2022.cogalex-1.7.pdf

Cite this

Liu, C., & Chersoni, E. (2022). Exploring Nominal Coercion in Semantic Spaces with Static and Contextualized Word Embeddings. In M. Zock, E. Chersoni, Y.-Y. Hsu, & E. Santus (Eds.), Proceedings of the Workshop on Cognitive Aspects of the Lexicon (CogALex-VII) (pp. 49–57). Association for Computational Linguistics, ACL Anthology. https://aclanthology.org/2022.cogalex-1.7.pdf

@inproceedings{299c3550d85049f2b240e898a3073b2c,

title = "Exploring Nominal Coercion in Semantic Spaces with Static and Contextualized Word Embeddings",

abstract = "The distinction between mass nouns and count nouns has a long history in formal semantics, and linguists have been trying to identify the semantic properties defining the two classes. However, they also recognized that both can undergo meaning shifts and be used in contexts of a different type, via nominal coercion. In this paper, we present an approach to measure the meaning shift in count-mass coercion in English that makes use of static and contextualized word embedding distance. Our results show that the coercion shifts are detected only by a small subset of the traditional word embedding models, and that the shifts detected by the contextualized embedding of BERT are more pronounced for mass nouns.",

author = "Chenxin Liu and Emmanuele Chersoni",

year = "2022",

month = nov,

language = "English",

isbn = "978-1-959429-01-2",

pages = "49–57",

editor = "Michael Zock and Emmanuele Chersoni and Yu-Yin Hsu and Santus, {Enrico }",

booktitle = "Proceedings of the Workshop on Cognitive Aspects of the Lexicon (CogALex-VII)",

publisher = "Association for Computational Linguistics, ACL Anthology",

note = "AACL-IJCNLP Workshop on Cognitive Aspects of the Lexicon 2022, CogALex 2022 ; Conference date: 20-11-2022 Through 20-11-2022",

}

Liu, C & Chersoni, E 2022, Exploring Nominal Coercion in Semantic Spaces with Static and Contextualized Word Embeddings. in M Zock, E Chersoni, Y-Y Hsu & E Santus (eds), Proceedings of the Workshop on Cognitive Aspects of the Lexicon (CogALex-VII). Association for Computational Linguistics, ACL Anthology, pp. 49–57, AACL-IJCNLP Workshop on Cognitive Aspects of the Lexicon 2022, 20/11/22. <https://aclanthology.org/2022.cogalex-1.7.pdf>

Exploring Nominal Coercion in Semantic Spaces with Static and Contextualized Word Embeddings. / Liu, Chenxin; Chersoni, Emmanuele.
Proceedings of the Workshop on Cognitive Aspects of the Lexicon (CogALex-VII). ed. / Michael Zock; Emmanuele Chersoni; Yu-Yin Hsu; Enrico Santus. Association for Computational Linguistics, ACL Anthology, 2022. p. 49–57.

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

TY - GEN

T1 - Exploring Nominal Coercion in Semantic Spaces with Static and Contextualized Word Embeddings

AU - Liu, Chenxin

AU - Chersoni, Emmanuele

PY - 2022/11

Y1 - 2022/11

N2 - The distinction between mass nouns and count nouns has a long history in formal semantics, and linguists have been trying to identify the semantic properties defining the two classes. However, they also recognized that both can undergo meaning shifts and be used in contexts of a different type, via nominal coercion. In this paper, we present an approach to measure the meaning shift in count-mass coercion in English that makes use of static and contextualized word embedding distance. Our results show that the coercion shifts are detected only by a small subset of the traditional word embedding models, and that the shifts detected by the contextualized embedding of BERT are more pronounced for mass nouns.

AB - The distinction between mass nouns and count nouns has a long history in formal semantics, and linguists have been trying to identify the semantic properties defining the two classes. However, they also recognized that both can undergo meaning shifts and be used in contexts of a different type, via nominal coercion. In this paper, we present an approach to measure the meaning shift in count-mass coercion in English that makes use of static and contextualized word embedding distance. Our results show that the coercion shifts are detected only by a small subset of the traditional word embedding models, and that the shifts detected by the contextualized embedding of BERT are more pronounced for mass nouns.

M3 - Conference article published in proceeding or book

SN - 978-1-959429-01-2

SP - 49

EP - 57

BT - Proceedings of the Workshop on Cognitive Aspects of the Lexicon (CogALex-VII)

A2 - Zock, Michael

A2 - Chersoni, Emmanuele

A2 - Hsu, Yu-Yin

A2 - Santus, Enrico

PB - Association for Computational Linguistics, ACL Anthology

T2 - AACL-IJCNLP Workshop on Cognitive Aspects of the Lexicon 2022

Y2 - 20 November 2022 through 20 November 2022

ER -

Exploring Nominal Coercion in Semantic Spaces with Static and Contextualized Word Embeddings

Abstract

Conference

More information

Fingerprint

Cite this