@inproceedings{gueudre-jose-2022-constraining,
title = "Constraining word alignments with posterior regularization for label transfer",
author = "Jose, Kevin and
Gueudre, Thomas",
editor = "Loukina, Anastassia and
Gangadharaiah, Rashmi and
Min, Bonan",
booktitle = "Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Industry Track",
month = jul,
year = "2022",
address = "Hybrid: Seattle, Washington + Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.naacl-industry.15",
doi = "10.18653/v1/2022.naacl-industry.15",
pages = "121--129",
abstract = "Unsupervised word alignments offer a lightweight and interpretable method to transfer labels from high- to low-resource languages, as long as semantically related words have the same label across languages. But such an assumption is often not true in industrial NLP pipelines, where multilingual annotation guidelines are complex and deviate from semantic consistency due to various factors (such as annotation difficulty, conflicting ontology, upcoming feature launches etc.);We address this difficulty by constraining the alignments models to remain consistent with both source and target annotation guidelines , leveraging posterior regularization and labeled examples. We illustrate the overall approach using IBM 2 (fast{\_}align) as a base model, and report results on both internal and external annotated datasets. We measure consistent accuracy improvements on the MultiATIS++ dataset over AWESoME, a popular transformer-based alignment model, in the label projection task ($+2.7\%$ at word-level and $+15\%$ at sentence-level), and show how even a small amount of target language annotations help substantially.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gueudre-jose-2022-constraining">
<titleInfo>
<title>Constraining word alignments with posterior regularization for label transfer</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kevin</namePart>
<namePart type="family">Jose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Thomas</namePart>
<namePart type="family">Gueudre</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Industry Track</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anastassia</namePart>
<namePart type="family">Loukina</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rashmi</namePart>
<namePart type="family">Gangadharaiah</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bonan</namePart>
<namePart type="family">Min</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hybrid: Seattle, Washington + Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Unsupervised word alignments offer a lightweight and interpretable method to transfer labels from high- to low-resource languages, as long as semantically related words have the same label across languages. But such an assumption is often not true in industrial NLP pipelines, where multilingual annotation guidelines are complex and deviate from semantic consistency due to various factors (such as annotation difficulty, conflicting ontology, upcoming feature launches etc.);We address this difficulty by constraining the alignments models to remain consistent with both source and target annotation guidelines , leveraging posterior regularization and labeled examples. We illustrate the overall approach using IBM 2 (fast_align) as a base model, and report results on both internal and external annotated datasets. We measure consistent accuracy improvements on the MultiATIS++ dataset over AWESoME, a popular transformer-based alignment model, in the label projection task (+2.7% at word-level and +15% at sentence-level), and show how even a small amount of target language annotations help substantially.</abstract>
<identifier type="citekey">gueudre-jose-2022-constraining</identifier>
<identifier type="doi">10.18653/v1/2022.naacl-industry.15</identifier>
<location>
<url>https://aclanthology.org/2022.naacl-industry.15</url>
</location>
<part>
<date>2022-07</date>
<extent unit="page">
<start>121</start>
<end>129</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Constraining word alignments with posterior regularization for label transfer
%A Jose, Kevin
%A Gueudre, Thomas
%Y Loukina, Anastassia
%Y Gangadharaiah, Rashmi
%Y Min, Bonan
%S Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Industry Track
%D 2022
%8 July
%I Association for Computational Linguistics
%C Hybrid: Seattle, Washington + Online
%F gueudre-jose-2022-constraining
%X Unsupervised word alignments offer a lightweight and interpretable method to transfer labels from high- to low-resource languages, as long as semantically related words have the same label across languages. But such an assumption is often not true in industrial NLP pipelines, where multilingual annotation guidelines are complex and deviate from semantic consistency due to various factors (such as annotation difficulty, conflicting ontology, upcoming feature launches etc.);We address this difficulty by constraining the alignments models to remain consistent with both source and target annotation guidelines , leveraging posterior regularization and labeled examples. We illustrate the overall approach using IBM 2 (fast_align) as a base model, and report results on both internal and external annotated datasets. We measure consistent accuracy improvements on the MultiATIS++ dataset over AWESoME, a popular transformer-based alignment model, in the label projection task (+2.7% at word-level and +15% at sentence-level), and show how even a small amount of target language annotations help substantially.
%R 10.18653/v1/2022.naacl-industry.15
%U https://aclanthology.org/2022.naacl-industry.15
%U https://doi.org/10.18653/v1/2022.naacl-industry.15
%P 121-129
Markdown (Informal)
[Constraining word alignments with posterior regularization for label transfer](https://aclanthology.org/2022.naacl-industry.15) (Jose & Gueudre, NAACL 2022)
ACL