@inproceedings{mohammadshahi-etal-2022-small,
title = "{SM}a{LL}-100: Introducing Shallow Multilingual Machine Translation Model for Low-Resource Languages",
author = "Mohammadshahi, Alireza and
Nikoulina, Vassilina and
Berard, Alexandre and
Brun, Caroline and
Henderson, James and
Besacier, Laurent",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.emnlp-main.571",
doi = "10.18653/v1/2022.emnlp-main.571",
pages = "8348--8359",
abstract = "In recent years, multilingual machine translation models have achieved promising performance on low-resource language pairs by sharing information between similar languages, thus enabling zero-shot translation. To overcome the {``}curse of multilinguality{''}, these models often opt for scaling up the number of parameters, which makes their use in resource-constrained environments challenging. We introduce SMaLL-100, a distilled version of the M2M-100(12B) model, a massively multilingual machine translation model covering 100 languages. We train SMaLL-100 with uniform sampling across all language pairs and therefore focus on preserving the performance of low-resource languages. We evaluate SMaLL-100 on different low-resource benchmarks: FLORES-101, Tatoeba, and TICO-19 and demonstrate that it outperforms previous massively multilingual models of comparable sizes (200-600M) while improving inference latency and memory usage. Additionally, our model achieves comparable results to M2M-100 (1.2B), while being 3.6x smaller and 4.3x faster at inference.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mohammadshahi-etal-2022-small">
<titleInfo>
<title>SMaLL-100: Introducing Shallow Multilingual Machine Translation Model for Low-Resource Languages</title>
</titleInfo>
<name type="personal">
<namePart type="given">Alireza</namePart>
<namePart type="family">Mohammadshahi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vassilina</namePart>
<namePart type="family">Nikoulina</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexandre</namePart>
<namePart type="family">Berard</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Caroline</namePart>
<namePart type="family">Brun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">James</namePart>
<namePart type="family">Henderson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Laurent</namePart>
<namePart type="family">Besacier</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yoav</namePart>
<namePart type="family">Goldberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zornitsa</namePart>
<namePart type="family">Kozareva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, United Arab Emirates</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>In recent years, multilingual machine translation models have achieved promising performance on low-resource language pairs by sharing information between similar languages, thus enabling zero-shot translation. To overcome the “curse of multilinguality”, these models often opt for scaling up the number of parameters, which makes their use in resource-constrained environments challenging. We introduce SMaLL-100, a distilled version of the M2M-100(12B) model, a massively multilingual machine translation model covering 100 languages. We train SMaLL-100 with uniform sampling across all language pairs and therefore focus on preserving the performance of low-resource languages. We evaluate SMaLL-100 on different low-resource benchmarks: FLORES-101, Tatoeba, and TICO-19 and demonstrate that it outperforms previous massively multilingual models of comparable sizes (200-600M) while improving inference latency and memory usage. Additionally, our model achieves comparable results to M2M-100 (1.2B), while being 3.6x smaller and 4.3x faster at inference.</abstract>
<identifier type="citekey">mohammadshahi-etal-2022-small</identifier>
<identifier type="doi">10.18653/v1/2022.emnlp-main.571</identifier>
<location>
<url>https://aclanthology.org/2022.emnlp-main.571</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>8348</start>
<end>8359</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SMaLL-100: Introducing Shallow Multilingual Machine Translation Model for Low-Resource Languages
%A Mohammadshahi, Alireza
%A Nikoulina, Vassilina
%A Berard, Alexandre
%A Brun, Caroline
%A Henderson, James
%A Besacier, Laurent
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F mohammadshahi-etal-2022-small
%X In recent years, multilingual machine translation models have achieved promising performance on low-resource language pairs by sharing information between similar languages, thus enabling zero-shot translation. To overcome the “curse of multilinguality”, these models often opt for scaling up the number of parameters, which makes their use in resource-constrained environments challenging. We introduce SMaLL-100, a distilled version of the M2M-100(12B) model, a massively multilingual machine translation model covering 100 languages. We train SMaLL-100 with uniform sampling across all language pairs and therefore focus on preserving the performance of low-resource languages. We evaluate SMaLL-100 on different low-resource benchmarks: FLORES-101, Tatoeba, and TICO-19 and demonstrate that it outperforms previous massively multilingual models of comparable sizes (200-600M) while improving inference latency and memory usage. Additionally, our model achieves comparable results to M2M-100 (1.2B), while being 3.6x smaller and 4.3x faster at inference.
%R 10.18653/v1/2022.emnlp-main.571
%U https://aclanthology.org/2022.emnlp-main.571
%U https://doi.org/10.18653/v1/2022.emnlp-main.571
%P 8348-8359
Markdown (Informal)
[SMaLL-100: Introducing Shallow Multilingual Machine Translation Model for Low-Resource Languages](https://aclanthology.org/2022.emnlp-main.571) (Mohammadshahi et al., EMNLP 2022)
ACL