@inproceedings{wu-dredze-2020-explicit,
title = "Do Explicit Alignments Robustly Improve Multilingual Encoders?",
author = "Wu, Shijie and
Dredze, Mark",
editor = "Webber, Bonnie and
Cohn, Trevor and
He, Yulan and
Liu, Yang",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.emnlp-main.362",
doi = "10.18653/v1/2020.emnlp-main.362",
pages = "4471--4482",
abstract = "Multilingual BERT (mBERT), XLM-RoBERTa (XLMR) and other unsupervised multilingual encoders can effectively learn cross-lingual representation. Explicit alignment objectives based on bitexts like Europarl or MultiUN have been shown to further improve these representations. However, word-level alignments are often suboptimal and such bitexts are unavailable for many languages. In this paper, we propose a new contrastive alignment objective that can better utilize such signal, and examine whether these previous alignment methods can be adapted to noisier sources of aligned data: a randomly sampled 1 million pair subset of the OPUS collection. Additionally, rather than report results on a single dataset with a single model run, we report the mean and standard derivation of multiple runs with different seeds, on four datasets and tasks. Our more extensive analysis finds that, while our new objective outperforms previous work, overall these methods do not improve performance with a more robust evaluation framework. Furthermore, the gains from using a better underlying model eclipse any benefits from alignment training. These negative results dictate more care in evaluating these methods and suggest limitations in applying explicit alignment objectives.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wu-dredze-2020-explicit">
<titleInfo>
<title>Do Explicit Alignments Robustly Improve Multilingual Encoders?</title>
</titleInfo>
<name type="personal">
<namePart type="given">Shijie</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mark</namePart>
<namePart type="family">Dredze</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Bonnie</namePart>
<namePart type="family">Webber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Trevor</namePart>
<namePart type="family">Cohn</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yulan</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Multilingual BERT (mBERT), XLM-RoBERTa (XLMR) and other unsupervised multilingual encoders can effectively learn cross-lingual representation. Explicit alignment objectives based on bitexts like Europarl or MultiUN have been shown to further improve these representations. However, word-level alignments are often suboptimal and such bitexts are unavailable for many languages. In this paper, we propose a new contrastive alignment objective that can better utilize such signal, and examine whether these previous alignment methods can be adapted to noisier sources of aligned data: a randomly sampled 1 million pair subset of the OPUS collection. Additionally, rather than report results on a single dataset with a single model run, we report the mean and standard derivation of multiple runs with different seeds, on four datasets and tasks. Our more extensive analysis finds that, while our new objective outperforms previous work, overall these methods do not improve performance with a more robust evaluation framework. Furthermore, the gains from using a better underlying model eclipse any benefits from alignment training. These negative results dictate more care in evaluating these methods and suggest limitations in applying explicit alignment objectives.</abstract>
<identifier type="citekey">wu-dredze-2020-explicit</identifier>
<identifier type="doi">10.18653/v1/2020.emnlp-main.362</identifier>
<location>
<url>https://aclanthology.org/2020.emnlp-main.362</url>
</location>
<part>
<date>2020-11</date>
<extent unit="page">
<start>4471</start>
<end>4482</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Do Explicit Alignments Robustly Improve Multilingual Encoders?
%A Wu, Shijie
%A Dredze, Mark
%Y Webber, Bonnie
%Y Cohn, Trevor
%Y He, Yulan
%Y Liu, Yang
%S Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
%D 2020
%8 November
%I Association for Computational Linguistics
%C Online
%F wu-dredze-2020-explicit
%X Multilingual BERT (mBERT), XLM-RoBERTa (XLMR) and other unsupervised multilingual encoders can effectively learn cross-lingual representation. Explicit alignment objectives based on bitexts like Europarl or MultiUN have been shown to further improve these representations. However, word-level alignments are often suboptimal and such bitexts are unavailable for many languages. In this paper, we propose a new contrastive alignment objective that can better utilize such signal, and examine whether these previous alignment methods can be adapted to noisier sources of aligned data: a randomly sampled 1 million pair subset of the OPUS collection. Additionally, rather than report results on a single dataset with a single model run, we report the mean and standard derivation of multiple runs with different seeds, on four datasets and tasks. Our more extensive analysis finds that, while our new objective outperforms previous work, overall these methods do not improve performance with a more robust evaluation framework. Furthermore, the gains from using a better underlying model eclipse any benefits from alignment training. These negative results dictate more care in evaluating these methods and suggest limitations in applying explicit alignment objectives.
%R 10.18653/v1/2020.emnlp-main.362
%U https://aclanthology.org/2020.emnlp-main.362
%U https://doi.org/10.18653/v1/2020.emnlp-main.362
%P 4471-4482
Markdown (Informal)
[Do Explicit Alignments Robustly Improve Multilingual Encoders?](https://aclanthology.org/2020.emnlp-main.362) (Wu & Dredze, EMNLP 2020)
ACL