Does Typological Blinding Impede Cross-Lingual Sharing?

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Standard

Does Typological Blinding Impede Cross-Lingual Sharing? / Bjerva, Johannes; Augenstein, Isabelle.

Proceedings of The 16th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2021). Association for Computational Linguistics, 2021.

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Harvard

Bjerva, J & Augenstein, I 2021, Does Typological Blinding Impede Cross-Lingual Sharing? in Proceedings of The 16th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2021). Association for Computational Linguistics, The 16th Conference of the European Chapter
of the Association for Computational Linguistics, 21/04/2021. <https://arxiv.org/abs/2101.11888>

APA

Bjerva, J., & Augenstein, I. (Accepted/In press). Does Typological Blinding Impede Cross-Lingual Sharing? In Proceedings of The 16th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2021) Association for Computational Linguistics. https://arxiv.org/abs/2101.11888

Vancouver

Bjerva J, Augenstein I. Does Typological Blinding Impede Cross-Lingual Sharing? In Proceedings of The 16th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2021). Association for Computational Linguistics. 2021

Author

Bjerva, Johannes ; Augenstein, Isabelle. / Does Typological Blinding Impede Cross-Lingual Sharing?. Proceedings of The 16th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2021). Association for Computational Linguistics, 2021.

Bibtex

@inproceedings{482614403c8d4b5daa36ef7f0c6c22ac,
title = "Does Typological Blinding Impede Cross-Lingual Sharing?",
abstract = "Bridging the performance gap between high- and low-resource languages has been the focus of much previous work. Typological features from databases such as the World Atlas of Language Structures (WALS) are a prime candidate for this, as such data exists even for very low-resource languages. However, previous work has only found minor benefits from using typological information. Our hypothesis is that a model trained in a cross-lingual setting will pick up on typological cues from the input data, thus overshadowing the utility of explicitly using such features. We verify this hypothesis by blinding a model to typological information, and investigate how cross-lingual sharing and performance is impacted. Our model is based on a cross-lingual architecture in which the latent weights governing the sharing between languages is learnt during training. We show that (i) preventing this model from exploiting typology severely reduces performance, while a control experiment reaffirms that (ii) encouraging sharing according to typology somewhat improves performance. ",
author = "Johannes Bjerva and Isabelle Augenstein",
year = "2021",
language = "Dansk",
booktitle = "Proceedings of The 16th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2021)",
publisher = "Association for Computational Linguistics",
note = "null ; Conference date: 21-04-2021 Through 23-04-2021",
url = "https://2021.eacl.org/",

}

RIS

TY - GEN

T1 - Does Typological Blinding Impede Cross-Lingual Sharing?

AU - Bjerva, Johannes

AU - Augenstein, Isabelle

N1 - Conference code: 16

PY - 2021

Y1 - 2021

N2 - Bridging the performance gap between high- and low-resource languages has been the focus of much previous work. Typological features from databases such as the World Atlas of Language Structures (WALS) are a prime candidate for this, as such data exists even for very low-resource languages. However, previous work has only found minor benefits from using typological information. Our hypothesis is that a model trained in a cross-lingual setting will pick up on typological cues from the input data, thus overshadowing the utility of explicitly using such features. We verify this hypothesis by blinding a model to typological information, and investigate how cross-lingual sharing and performance is impacted. Our model is based on a cross-lingual architecture in which the latent weights governing the sharing between languages is learnt during training. We show that (i) preventing this model from exploiting typology severely reduces performance, while a control experiment reaffirms that (ii) encouraging sharing according to typology somewhat improves performance.

AB - Bridging the performance gap between high- and low-resource languages has been the focus of much previous work. Typological features from databases such as the World Atlas of Language Structures (WALS) are a prime candidate for this, as such data exists even for very low-resource languages. However, previous work has only found minor benefits from using typological information. Our hypothesis is that a model trained in a cross-lingual setting will pick up on typological cues from the input data, thus overshadowing the utility of explicitly using such features. We verify this hypothesis by blinding a model to typological information, and investigate how cross-lingual sharing and performance is impacted. Our model is based on a cross-lingual architecture in which the latent weights governing the sharing between languages is learnt during training. We show that (i) preventing this model from exploiting typology severely reduces performance, while a control experiment reaffirms that (ii) encouraging sharing according to typology somewhat improves performance.

M3 - Konferencebidrag i proceedings

BT - Proceedings of The 16th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2021)

PB - Association for Computational Linguistics

Y2 - 21 April 2021 through 23 April 2021

ER -

ID: 260407759