@inproceedings{agirrezabal-etal-2023-hidden,
title = "The Hidden Folk: Linguistic Properties Encoded in Multilingual Contextual Character Representations",
author = "Agirrezabal, Manex and
Boldsen, Sidsel and
Hollenstein, Nora",
editor = "Gorman, Kyle and
Sproat, Richard and
Roark, Brian",
booktitle = "Proceedings of the Workshop on Computation and Written Language (CAWL 2023)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.cawl-1.2",
doi = "10.18653/v1/2023.cawl-1.2",
pages = "6--13",
abstract = "To gain a better understanding of the linguistic information encoded in character-based language models, we probe the multilingual contextual CANINE model. We design a range of phonetic probing tasks in six Nordic languages, including Faroese as an additional zero-shot instance. We observe that some phonetic information is indeed encoded in the character representations, as consonants and vowels can be well distinguished using a linear classifier. Furthermore, results for the Danish and Norwegian language seem to be worse for the consonant/vowel distinction in comparison to other languages. The information encoded in these representations can also be learned in a zero-shot scenario, as Faroese shows a reasonably good performance in the same vowel/consonant distinction task.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="agirrezabal-etal-2023-hidden">
<titleInfo>
<title>The Hidden Folk: Linguistic Properties Encoded in Multilingual Contextual Character Representations</title>
</titleInfo>
<name type="personal">
<namePart type="given">Manex</namePart>
<namePart type="family">Agirrezabal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sidsel</namePart>
<namePart type="family">Boldsen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nora</namePart>
<namePart type="family">Hollenstein</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Workshop on Computation and Written Language (CAWL 2023)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kyle</namePart>
<namePart type="family">Gorman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Richard</namePart>
<namePart type="family">Sproat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Brian</namePart>
<namePart type="family">Roark</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>To gain a better understanding of the linguistic information encoded in character-based language models, we probe the multilingual contextual CANINE model. We design a range of phonetic probing tasks in six Nordic languages, including Faroese as an additional zero-shot instance. We observe that some phonetic information is indeed encoded in the character representations, as consonants and vowels can be well distinguished using a linear classifier. Furthermore, results for the Danish and Norwegian language seem to be worse for the consonant/vowel distinction in comparison to other languages. The information encoded in these representations can also be learned in a zero-shot scenario, as Faroese shows a reasonably good performance in the same vowel/consonant distinction task.</abstract>
<identifier type="citekey">agirrezabal-etal-2023-hidden</identifier>
<identifier type="doi">10.18653/v1/2023.cawl-1.2</identifier>
<location>
<url>https://aclanthology.org/2023.cawl-1.2</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>6</start>
<end>13</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T The Hidden Folk: Linguistic Properties Encoded in Multilingual Contextual Character Representations
%A Agirrezabal, Manex
%A Boldsen, Sidsel
%A Hollenstein, Nora
%Y Gorman, Kyle
%Y Sproat, Richard
%Y Roark, Brian
%S Proceedings of the Workshop on Computation and Written Language (CAWL 2023)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F agirrezabal-etal-2023-hidden
%X To gain a better understanding of the linguistic information encoded in character-based language models, we probe the multilingual contextual CANINE model. We design a range of phonetic probing tasks in six Nordic languages, including Faroese as an additional zero-shot instance. We observe that some phonetic information is indeed encoded in the character representations, as consonants and vowels can be well distinguished using a linear classifier. Furthermore, results for the Danish and Norwegian language seem to be worse for the consonant/vowel distinction in comparison to other languages. The information encoded in these representations can also be learned in a zero-shot scenario, as Faroese shows a reasonably good performance in the same vowel/consonant distinction task.
%R 10.18653/v1/2023.cawl-1.2
%U https://aclanthology.org/2023.cawl-1.2
%U https://doi.org/10.18653/v1/2023.cawl-1.2
%P 6-13
Markdown (Informal)
[The Hidden Folk: Linguistic Properties Encoded in Multilingual Contextual Character Representations](https://aclanthology.org/2023.cawl-1.2) (Agirrezabal et al., CAWL 2023)
ACL