@article{ustun-etal-2022-udapter,
title = "{UD}apter: Typology-based Language Adapters for Multilingual Dependency Parsing and Sequence Labeling",
author = {{\"U}st{\"u}n, Ahmet and
Bisazza, Arianna and
Bouma, Gosse and
van Noord, Gertjan},
journal = "Computational Linguistics",
volume = "48",
number = "3",
month = sep,
year = "2022",
address = "Cambridge, MA",
publisher = "MIT Press",
url = "https://aclanthology.org/2022.cl-3.3",
doi = "10.1162/coli_a_00443",
pages = "555--592",
abstract = "Recent advances in multilingual language modeling have brought the idea of a truly universal parser closer to reality. However, such models are still not immune to the {``}curse of multilinguality{''}: Cross-language interference and restrained model capacity remain major obstacles. To address this, we propose a novel language adaptation approach by introducing contextual language adapters to a multilingual parser. Contextual language adapters make it possible to learn adapters via language embeddings while sharing model parameters across languages based on contextual parameter generation. Moreover, our method allows for an easy but effective integration of existing linguistic typology features into the parsing model. Because not all typological features are available for every language, we further combine typological feature prediction with parsing in a multi-task model that achieves very competitive parsing performance without the need for an external prediction system for missing features. The resulting parser, UDapter, can be used for dependency parsing as well as sequence labeling tasks such as POS tagging, morphological tagging, and NER. In dependency parsing, it outperforms strong monolingual and multilingual baselines on the majority of both high-resource and low-resource (zero-shot) languages, showing the success of the proposed adaptation approach. In sequence labeling tasks, our parser surpasses the baseline on high resource languages, and performs very competitively in a zero-shot setting. Our in-depth analyses show that adapter generation via typological features of languages is key to this success.1",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ustun-etal-2022-udapter">
<titleInfo>
<title>UDapter: Typology-based Language Adapters for Multilingual Dependency Parsing and Sequence Labeling</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ahmet</namePart>
<namePart type="family">Üstün</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Arianna</namePart>
<namePart type="family">Bisazza</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gosse</namePart>
<namePart type="family">Bouma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gertjan</namePart>
<namePart type="family">van Noord</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-09</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<genre authority="bibutilsgt">journal article</genre>
<relatedItem type="host">
<titleInfo>
<title>Computational Linguistics</title>
</titleInfo>
<originInfo>
<issuance>continuing</issuance>
<publisher>MIT Press</publisher>
<place>
<placeTerm type="text">Cambridge, MA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">periodical</genre>
<genre authority="bibutilsgt">academic journal</genre>
</relatedItem>
<abstract>Recent advances in multilingual language modeling have brought the idea of a truly universal parser closer to reality. However, such models are still not immune to the “curse of multilinguality”: Cross-language interference and restrained model capacity remain major obstacles. To address this, we propose a novel language adaptation approach by introducing contextual language adapters to a multilingual parser. Contextual language adapters make it possible to learn adapters via language embeddings while sharing model parameters across languages based on contextual parameter generation. Moreover, our method allows for an easy but effective integration of existing linguistic typology features into the parsing model. Because not all typological features are available for every language, we further combine typological feature prediction with parsing in a multi-task model that achieves very competitive parsing performance without the need for an external prediction system for missing features. The resulting parser, UDapter, can be used for dependency parsing as well as sequence labeling tasks such as POS tagging, morphological tagging, and NER. In dependency parsing, it outperforms strong monolingual and multilingual baselines on the majority of both high-resource and low-resource (zero-shot) languages, showing the success of the proposed adaptation approach. In sequence labeling tasks, our parser surpasses the baseline on high resource languages, and performs very competitively in a zero-shot setting. Our in-depth analyses show that adapter generation via typological features of languages is key to this success.1</abstract>
<identifier type="citekey">ustun-etal-2022-udapter</identifier>
<identifier type="doi">10.1162/coli_a_00443</identifier>
<location>
<url>https://aclanthology.org/2022.cl-3.3</url>
</location>
<part>
<date>2022-09</date>
<detail type="volume"><number>48</number></detail>
<detail type="issue"><number>3</number></detail>
<extent unit="page">
<start>555</start>
<end>592</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Journal Article
%T UDapter: Typology-based Language Adapters for Multilingual Dependency Parsing and Sequence Labeling
%A Üstün, Ahmet
%A Bisazza, Arianna
%A Bouma, Gosse
%A van Noord, Gertjan
%J Computational Linguistics
%D 2022
%8 September
%V 48
%N 3
%I MIT Press
%C Cambridge, MA
%F ustun-etal-2022-udapter
%X Recent advances in multilingual language modeling have brought the idea of a truly universal parser closer to reality. However, such models are still not immune to the “curse of multilinguality”: Cross-language interference and restrained model capacity remain major obstacles. To address this, we propose a novel language adaptation approach by introducing contextual language adapters to a multilingual parser. Contextual language adapters make it possible to learn adapters via language embeddings while sharing model parameters across languages based on contextual parameter generation. Moreover, our method allows for an easy but effective integration of existing linguistic typology features into the parsing model. Because not all typological features are available for every language, we further combine typological feature prediction with parsing in a multi-task model that achieves very competitive parsing performance without the need for an external prediction system for missing features. The resulting parser, UDapter, can be used for dependency parsing as well as sequence labeling tasks such as POS tagging, morphological tagging, and NER. In dependency parsing, it outperforms strong monolingual and multilingual baselines on the majority of both high-resource and low-resource (zero-shot) languages, showing the success of the proposed adaptation approach. In sequence labeling tasks, our parser surpasses the baseline on high resource languages, and performs very competitively in a zero-shot setting. Our in-depth analyses show that adapter generation via typological features of languages is key to this success.1
%R 10.1162/coli_a_00443
%U https://aclanthology.org/2022.cl-3.3
%U https://doi.org/10.1162/coli_a_00443
%P 555-592
Markdown (Informal)
[UDapter: Typology-based Language Adapters for Multilingual Dependency Parsing and Sequence Labeling](https://aclanthology.org/2022.cl-3.3) (Üstün et al., CL 2022)
ACL