UoB_UK at SemEval 2021 task 2: zero-shot and few-shot learning for multi-lingual and cross-lingual word sense disambiguation

Wei Li; Harish Tayyar Madabushi; Mark Lee

doi:10.18653/v1/2021.semeval-1.97

UoB_UK at SemEval 2021 task 2: zero-shot and few-shot learning for multi-lingual and cross-lingual word sense disambiguation

Wei Li, Harish Tayyar Madabushi, Mark Lee

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

11 Downloads (Pure)

Abstract

This paper describes our submission to SemEval 2021 Task 2. We compare XLM-RoBERTa Base and Large in the few-shot and zero-shot settings and additionally test the effectiveness of using a k-nearest neighbors classifier in the few-shot setting instead of the more traditional multi-layered perceptron. Our experiments on both the multi-lingual and cross-lingual data show that XLM-RoBERTa Large, unlike the Base version, seems to be able to more effectively transfer learning in a few-shot setting and that the k-nearest neighbors classifier is indeed a more powerful classifier than a multi-layered perceptron when used in few-shot learning.

Original language	English
Title of host publication	Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021)
Editors	Alexis Palmer, Nathan Schneider, Natalie Schluter, Guy Emerson, Aurelie Herbelot, Xiaodan Zhu
Publisher	Association for Computational Linguistics, ACL
Pages	738-742
Number of pages	5
ISBN (Electronic)	9781954085701
ISBN (Print)	9781954085701
DOIs	https://doi.org/10.18653/v1/2021.semeval-1.97
Publication status	Published - Aug 2021
Event	15th International Workshop on Semantic Evaluation, SemEval 2021 - Virtual, Bangkok, Thailand Duration: 5 Aug 2021 → 6 Aug 2021

Publication series

Name	Lexical and Computational Semantics and Semantic Evaluation (formerly Workshop on Sense Evaluation) (SemEval)

Conference

Conference	15th International Workshop on Semantic Evaluation, SemEval 2021
Country/Territory	Thailand
City	Virtual, Bangkok
Period	5/08/21 → 6/08/21

Bibliographical note

Publisher Copyright:
© 2021 Association for Computational Linguistics.

ASJC Scopus subject areas

Computational Theory and Mathematics
Computer Science Applications
Theoretical Computer Science

Access to Document

10.18653/v1/2021.semeval-1.97Licence: Creative Commons: Attribution (CC BY)

LiW2021UoBFinal published version, 262 KBLicence: Creative Commons: Attribution (CC BY)

Cite this

Li, W., Madabushi, H. T., & Lee, M. (2021). UoB_UK at SemEval 2021 task 2: zero-shot and few-shot learning for multi-lingual and cross-lingual word sense disambiguation. In A. Palmer, N. Schneider, N. Schluter, G. Emerson, A. Herbelot, & X. Zhu (Eds.), Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021) (pp. 738-742). (Lexical and Computational Semantics and Semantic Evaluation (formerly Workshop on Sense Evaluation) (SemEval)). Association for Computational Linguistics, ACL. https://doi.org/10.18653/v1/2021.semeval-1.97

Li, Wei ; Madabushi, Harish Tayyar ; Lee, Mark. / UoB_UK at SemEval 2021 task 2 : zero-shot and few-shot learning for multi-lingual and cross-lingual word sense disambiguation. Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021). editor / Alexis Palmer ; Nathan Schneider ; Natalie Schluter ; Guy Emerson ; Aurelie Herbelot ; Xiaodan Zhu. Association for Computational Linguistics, ACL, 2021. pp. 738-742 (Lexical and Computational Semantics and Semantic Evaluation (formerly Workshop on Sense Evaluation) (SemEval)).

@inproceedings{64fc4e97956b4b359b3a84de1ef13109,

title = "UoB_UK at SemEval 2021 task 2: zero-shot and few-shot learning for multi-lingual and cross-lingual word sense disambiguation",

abstract = "This paper describes our submission to SemEval 2021 Task 2. We compare XLM-RoBERTa Base and Large in the few-shot and zero-shot settings and additionally test the effectiveness of using a k-nearest neighbors classifier in the few-shot setting instead of the more traditional multi-layered perceptron. Our experiments on both the multi-lingual and cross-lingual data show that XLM-RoBERTa Large, unlike the Base version, seems to be able to more effectively transfer learning in a few-shot setting and that the k-nearest neighbors classifier is indeed a more powerful classifier than a multi-layered perceptron when used in few-shot learning.",

author = "Wei Li and Madabushi, {Harish Tayyar} and Mark Lee",

note = "Publisher Copyright: {\textcopyright} 2021 Association for Computational Linguistics.; 15th International Workshop on Semantic Evaluation, SemEval 2021 ; Conference date: 05-08-2021 Through 06-08-2021",

year = "2021",

month = aug,

doi = "10.18653/v1/2021.semeval-1.97",

language = "English",

isbn = "9781954085701",

series = "Lexical and Computational Semantics and Semantic Evaluation (formerly Workshop on Sense Evaluation) (SemEval)",

publisher = "Association for Computational Linguistics, ACL",

pages = "738--742",

editor = "Alexis Palmer and Nathan Schneider and Natalie Schluter and Guy Emerson and Aurelie Herbelot and Xiaodan Zhu",

booktitle = "Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021)",

}

Li, W, Madabushi, HT & Lee, M 2021, UoB_UK at SemEval 2021 task 2: zero-shot and few-shot learning for multi-lingual and cross-lingual word sense disambiguation. in A Palmer, N Schneider, N Schluter, G Emerson, A Herbelot & X Zhu (eds), Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021). Lexical and Computational Semantics and Semantic Evaluation (formerly Workshop on Sense Evaluation) (SemEval), Association for Computational Linguistics, ACL, pp. 738-742, 15th International Workshop on Semantic Evaluation, SemEval 2021, Virtual, Bangkok, Thailand, 5/08/21. https://doi.org/10.18653/v1/2021.semeval-1.97

UoB_UK at SemEval 2021 task 2: zero-shot and few-shot learning for multi-lingual and cross-lingual word sense disambiguation. / Li, Wei; Madabushi, Harish Tayyar; Lee, Mark.
Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021). ed. / Alexis Palmer; Nathan Schneider; Natalie Schluter; Guy Emerson; Aurelie Herbelot; Xiaodan Zhu. Association for Computational Linguistics, ACL, 2021. p. 738-742 (Lexical and Computational Semantics and Semantic Evaluation (formerly Workshop on Sense Evaluation) (SemEval)).

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

TY - GEN

T1 - UoB_UK at SemEval 2021 task 2

T2 - 15th International Workshop on Semantic Evaluation, SemEval 2021

AU - Li, Wei

AU - Madabushi, Harish Tayyar

AU - Lee, Mark

PY - 2021/8

Y1 - 2021/8

N2 - This paper describes our submission to SemEval 2021 Task 2. We compare XLM-RoBERTa Base and Large in the few-shot and zero-shot settings and additionally test the effectiveness of using a k-nearest neighbors classifier in the few-shot setting instead of the more traditional multi-layered perceptron. Our experiments on both the multi-lingual and cross-lingual data show that XLM-RoBERTa Large, unlike the Base version, seems to be able to more effectively transfer learning in a few-shot setting and that the k-nearest neighbors classifier is indeed a more powerful classifier than a multi-layered perceptron when used in few-shot learning.

AB - This paper describes our submission to SemEval 2021 Task 2. We compare XLM-RoBERTa Base and Large in the few-shot and zero-shot settings and additionally test the effectiveness of using a k-nearest neighbors classifier in the few-shot setting instead of the more traditional multi-layered perceptron. Our experiments on both the multi-lingual and cross-lingual data show that XLM-RoBERTa Large, unlike the Base version, seems to be able to more effectively transfer learning in a few-shot setting and that the k-nearest neighbors classifier is indeed a more powerful classifier than a multi-layered perceptron when used in few-shot learning.

UR - http://www.scopus.com/inward/record.url?scp=85138931782&partnerID=8YFLogxK

U2 - 10.18653/v1/2021.semeval-1.97

DO - 10.18653/v1/2021.semeval-1.97

M3 - Conference contribution

AN - SCOPUS:85138931782

SN - 9781954085701

T3 - Lexical and Computational Semantics and Semantic Evaluation (formerly Workshop on Sense Evaluation) (SemEval)

SP - 738

EP - 742

BT - Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021)

A2 - Palmer, Alexis

A2 - Schneider, Nathan

A2 - Schluter, Natalie

A2 - Emerson, Guy

A2 - Herbelot, Aurelie

A2 - Zhu, Xiaodan

PB - Association for Computational Linguistics, ACL

Y2 - 5 August 2021 through 6 August 2021

ER -

Li W, Madabushi HT, Lee M. UoB_UK at SemEval 2021 task 2: zero-shot and few-shot learning for multi-lingual and cross-lingual word sense disambiguation. In Palmer A, Schneider N, Schluter N, Emerson G, Herbelot A, Zhu X, editors, Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021). Association for Computational Linguistics, ACL. 2021. p. 738-742. (Lexical and Computational Semantics and Semantic Evaluation (formerly Workshop on Sense Evaluation) (SemEval)). doi: 10.18653/v1/2021.semeval-1.97

UoB_UK at SemEval 2021 task 2: zero-shot and few-shot learning for multi-lingual and cross-lingual word sense disambiguation

Abstract

Publication series

Conference

Bibliographical note

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this