@inproceedings{huang-etal-2024-unsupervised,
title = "Unsupervised Multilingual Dense Retrieval via Generative Pseudo Labeling",
author = "Huang, Chao-Wei and
Li, Chen-An and
Hsu, Tsu-Yuan and
Hsu, Chen-Yu and
Chen, Yun-Nung",
editor = "Graham, Yvette and
Purver, Matthew",
booktitle = "Findings of the Association for Computational Linguistics: EACL 2024",
month = mar,
year = "2024",
address = "St. Julian{'}s, Malta",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-eacl.49",
pages = "736--746",
abstract = "Dense retrieval methods have demonstrated promising performance in multilingual information retrieval, where queries and documents can be in different languages. However, dense retrievers typically require a substantial amount of paired data, which poses even greater challenges in multilingual scenarios. This paper introduces $\textbf{UMR}$, an $\underline{U}$nsupervised $\underline{M}$ultilingual dense $\underline{R}$etriever trained without any paired data. Our approach leverages the sequence likelihood estimation capabilities of multilingual language models to acquire pseudo labels for training dense retrievers. We propose a two-stage framework which iteratively improves the performance of multilingual dense retrievers. Experimental results on two benchmark datasets show that UMR outperforms supervised baselines, showcasing the potential of training multilingual retrievers without paired data, thereby enhancing their practicality. All of our source code, data, and models are available: https://github.com/MiuLab/UMR",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="huang-etal-2024-unsupervised">
<titleInfo>
<title>Unsupervised Multilingual Dense Retrieval via Generative Pseudo Labeling</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chao-Wei</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chen-An</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tsu-Yuan</namePart>
<namePart type="family">Hsu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chen-Yu</namePart>
<namePart type="family">Hsu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yun-Nung</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-03</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EACL 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yvette</namePart>
<namePart type="family">Graham</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Matthew</namePart>
<namePart type="family">Purver</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">St. Julian’s, Malta</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Dense retrieval methods have demonstrated promising performance in multilingual information retrieval, where queries and documents can be in different languages. However, dense retrievers typically require a substantial amount of paired data, which poses even greater challenges in multilingual scenarios. This paper introduces UMR, an \underlineUnsupervised \underlineMultilingual dense \underlineRetriever trained without any paired data. Our approach leverages the sequence likelihood estimation capabilities of multilingual language models to acquire pseudo labels for training dense retrievers. We propose a two-stage framework which iteratively improves the performance of multilingual dense retrievers. Experimental results on two benchmark datasets show that UMR outperforms supervised baselines, showcasing the potential of training multilingual retrievers without paired data, thereby enhancing their practicality. All of our source code, data, and models are available: https://github.com/MiuLab/UMR</abstract>
<identifier type="citekey">huang-etal-2024-unsupervised</identifier>
<location>
<url>https://aclanthology.org/2024.findings-eacl.49</url>
</location>
<part>
<date>2024-03</date>
<extent unit="page">
<start>736</start>
<end>746</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Unsupervised Multilingual Dense Retrieval via Generative Pseudo Labeling
%A Huang, Chao-Wei
%A Li, Chen-An
%A Hsu, Tsu-Yuan
%A Hsu, Chen-Yu
%A Chen, Yun-Nung
%Y Graham, Yvette
%Y Purver, Matthew
%S Findings of the Association for Computational Linguistics: EACL 2024
%D 2024
%8 March
%I Association for Computational Linguistics
%C St. Julian’s, Malta
%F huang-etal-2024-unsupervised
%X Dense retrieval methods have demonstrated promising performance in multilingual information retrieval, where queries and documents can be in different languages. However, dense retrievers typically require a substantial amount of paired data, which poses even greater challenges in multilingual scenarios. This paper introduces UMR, an \underlineUnsupervised \underlineMultilingual dense \underlineRetriever trained without any paired data. Our approach leverages the sequence likelihood estimation capabilities of multilingual language models to acquire pseudo labels for training dense retrievers. We propose a two-stage framework which iteratively improves the performance of multilingual dense retrievers. Experimental results on two benchmark datasets show that UMR outperforms supervised baselines, showcasing the potential of training multilingual retrievers without paired data, thereby enhancing their practicality. All of our source code, data, and models are available: https://github.com/MiuLab/UMR
%U https://aclanthology.org/2024.findings-eacl.49
%P 736-746
Markdown (Informal)
[Unsupervised Multilingual Dense Retrieval via Generative Pseudo Labeling](https://aclanthology.org/2024.findings-eacl.49) (Huang et al., Findings 2024)
ACL