@inproceedings{mullov-waibel-2025-shot,
title = "Few-Shot Learning Translation from New Languages",
author = "Mullov, Carlos and
Waibel, Alexander",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.emnlp-main.163/",
pages = "3309--3330",
ISBN = "979-8-89176-332-6",
abstract = "Recent work shows strong transfer learning capability to unseen languages in sequence-to-sequence neural networks, under the assumption that we have high-quality word representations for the target language. We evaluate whether this direction is a viable path forward for translation from low-resource languages by investigating how much data is required to learn such high-quality word representations. We first show that learning word embeddings separately from a translation model can enable rapid adaptation to new languages with only a few hundred sentences of parallel data. To see whether the current bottleneck in transfer to low-resource languages lies mainly with learning the word representations, we then train word embeddings models on varying amounts of data, to then plug them into a machine translation model. We show that in this simulated low-resource setting with only 500 parallel sentences and 31,250 sentences of monolingual data we can exceed 15 BLEU on Flores on unseen languages. Finally, we investigate why on a real low-resource language the results are less favorable and find fault with the publicly available multilingual language modelling datasets."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mullov-waibel-2025-shot">
<titleInfo>
<title>Few-Shot Learning Translation from New Languages</title>
</titleInfo>
<name type="personal">
<namePart type="given">Carlos</namePart>
<namePart type="family">Mullov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexander</namePart>
<namePart type="family">Waibel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Christodoulopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolyn</namePart>
<namePart type="family">Rose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Violet</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-332-6</identifier>
</relatedItem>
<abstract>Recent work shows strong transfer learning capability to unseen languages in sequence-to-sequence neural networks, under the assumption that we have high-quality word representations for the target language. We evaluate whether this direction is a viable path forward for translation from low-resource languages by investigating how much data is required to learn such high-quality word representations. We first show that learning word embeddings separately from a translation model can enable rapid adaptation to new languages with only a few hundred sentences of parallel data. To see whether the current bottleneck in transfer to low-resource languages lies mainly with learning the word representations, we then train word embeddings models on varying amounts of data, to then plug them into a machine translation model. We show that in this simulated low-resource setting with only 500 parallel sentences and 31,250 sentences of monolingual data we can exceed 15 BLEU on Flores on unseen languages. Finally, we investigate why on a real low-resource language the results are less favorable and find fault with the publicly available multilingual language modelling datasets.</abstract>
<identifier type="citekey">mullov-waibel-2025-shot</identifier>
<location>
<url>https://aclanthology.org/2025.emnlp-main.163/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>3309</start>
<end>3330</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Few-Shot Learning Translation from New Languages
%A Mullov, Carlos
%A Waibel, Alexander
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-332-6
%F mullov-waibel-2025-shot
%X Recent work shows strong transfer learning capability to unseen languages in sequence-to-sequence neural networks, under the assumption that we have high-quality word representations for the target language. We evaluate whether this direction is a viable path forward for translation from low-resource languages by investigating how much data is required to learn such high-quality word representations. We first show that learning word embeddings separately from a translation model can enable rapid adaptation to new languages with only a few hundred sentences of parallel data. To see whether the current bottleneck in transfer to low-resource languages lies mainly with learning the word representations, we then train word embeddings models on varying amounts of data, to then plug them into a machine translation model. We show that in this simulated low-resource setting with only 500 parallel sentences and 31,250 sentences of monolingual data we can exceed 15 BLEU on Flores on unseen languages. Finally, we investigate why on a real low-resource language the results are less favorable and find fault with the publicly available multilingual language modelling datasets.
%U https://aclanthology.org/2025.emnlp-main.163/
%P 3309-3330
Markdown (Informal)
[Few-Shot Learning Translation from New Languages](https://aclanthology.org/2025.emnlp-main.163/) (Mullov & Waibel, EMNLP 2025)
ACL
- Carlos Mullov and Alexander Waibel. 2025. Few-Shot Learning Translation from New Languages. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 3309–3330, Suzhou, China. Association for Computational Linguistics.