@inproceedings{wang-etal-2021-efficient-test,
title = "Efficient Test Time Adapter Ensembling for Low-resource Language Varieties",
author = "Wang, Xinyi and
Tsvetkov, Yulia and
Ruder, Sebastian and
Neubig, Graham",
editor = "Moens, Marie-Francine and
Huang, Xuanjing and
Specia, Lucia and
Yih, Scott Wen-tau",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2021",
month = nov,
year = "2021",
address = "Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.findings-emnlp.63",
doi = "10.18653/v1/2021.findings-emnlp.63",
pages = "730--737",
abstract = "Adapters are light-weight modules that allow parameter-efficient fine-tuning of pretrained models. Specialized language and task adapters have recently been proposed to facilitate cross-lingual transfer of multilingual pretrained models (Pfeiffer et al., 2020b). However, this approach requires training a separate language adapter for every language one wishes to support, which can be impractical for languages with limited data. An intuitive solution is to use a related language adapter for the new language variety, but we observe that this solution can lead to sub-optimal performance. In this paper, we aim to improve the robustness of language adapters to uncovered languages without training new adapters. We find that ensembling multiple existing language adapters makes the fine-tuned model significantly more robust to other language varieties not included in these adapters. Building upon this observation, we propose Entropy Minimized Ensemble of Adapters (EMEA), a method that optimizes the ensemble weights of the pretrained language adapters for each test sentence by minimizing the entropy of its predictions. Experiments on three diverse groups of language varieties show that our method leads to significant improvements on both named entity recognition and part-of-speech tagging across all languages.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2021-efficient-test">
<titleInfo>
<title>Efficient Test Time Adapter Ensembling for Low-resource Language Varieties</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xinyi</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yulia</namePart>
<namePart type="family">Tsvetkov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sebastian</namePart>
<namePart type="family">Ruder</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Graham</namePart>
<namePart type="family">Neubig</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2021</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marie-Francine</namePart>
<namePart type="family">Moens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuanjing</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lucia</namePart>
<namePart type="family">Specia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Scott</namePart>
<namePart type="given">Wen-tau</namePart>
<namePart type="family">Yih</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Punta Cana, Dominican Republic</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Adapters are light-weight modules that allow parameter-efficient fine-tuning of pretrained models. Specialized language and task adapters have recently been proposed to facilitate cross-lingual transfer of multilingual pretrained models (Pfeiffer et al., 2020b). However, this approach requires training a separate language adapter for every language one wishes to support, which can be impractical for languages with limited data. An intuitive solution is to use a related language adapter for the new language variety, but we observe that this solution can lead to sub-optimal performance. In this paper, we aim to improve the robustness of language adapters to uncovered languages without training new adapters. We find that ensembling multiple existing language adapters makes the fine-tuned model significantly more robust to other language varieties not included in these adapters. Building upon this observation, we propose Entropy Minimized Ensemble of Adapters (EMEA), a method that optimizes the ensemble weights of the pretrained language adapters for each test sentence by minimizing the entropy of its predictions. Experiments on three diverse groups of language varieties show that our method leads to significant improvements on both named entity recognition and part-of-speech tagging across all languages.</abstract>
<identifier type="citekey">wang-etal-2021-efficient-test</identifier>
<identifier type="doi">10.18653/v1/2021.findings-emnlp.63</identifier>
<location>
<url>https://aclanthology.org/2021.findings-emnlp.63</url>
</location>
<part>
<date>2021-11</date>
<extent unit="page">
<start>730</start>
<end>737</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Efficient Test Time Adapter Ensembling for Low-resource Language Varieties
%A Wang, Xinyi
%A Tsvetkov, Yulia
%A Ruder, Sebastian
%A Neubig, Graham
%Y Moens, Marie-Francine
%Y Huang, Xuanjing
%Y Specia, Lucia
%Y Yih, Scott Wen-tau
%S Findings of the Association for Computational Linguistics: EMNLP 2021
%D 2021
%8 November
%I Association for Computational Linguistics
%C Punta Cana, Dominican Republic
%F wang-etal-2021-efficient-test
%X Adapters are light-weight modules that allow parameter-efficient fine-tuning of pretrained models. Specialized language and task adapters have recently been proposed to facilitate cross-lingual transfer of multilingual pretrained models (Pfeiffer et al., 2020b). However, this approach requires training a separate language adapter for every language one wishes to support, which can be impractical for languages with limited data. An intuitive solution is to use a related language adapter for the new language variety, but we observe that this solution can lead to sub-optimal performance. In this paper, we aim to improve the robustness of language adapters to uncovered languages without training new adapters. We find that ensembling multiple existing language adapters makes the fine-tuned model significantly more robust to other language varieties not included in these adapters. Building upon this observation, we propose Entropy Minimized Ensemble of Adapters (EMEA), a method that optimizes the ensemble weights of the pretrained language adapters for each test sentence by minimizing the entropy of its predictions. Experiments on three diverse groups of language varieties show that our method leads to significant improvements on both named entity recognition and part-of-speech tagging across all languages.
%R 10.18653/v1/2021.findings-emnlp.63
%U https://aclanthology.org/2021.findings-emnlp.63
%U https://doi.org/10.18653/v1/2021.findings-emnlp.63
%P 730-737
Markdown (Informal)
[Efficient Test Time Adapter Ensembling for Low-resource Language Varieties](https://aclanthology.org/2021.findings-emnlp.63) (Wang et al., Findings 2021)
ACL
