@inproceedings{jang-etal-2026-shift,
title = "{SHIFT}: Semantic Harmonization via Index-side Feature Transformation for Multilingual Information Retrieval",
author = "Jang, Youngjoon and
Hong, Seongtae and
Moon, Hyeonseok and
Lim, Heuiseok",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.109/",
pages = "2247--2269",
ISBN = "979-8-89176-462-0",
abstract = "With the rapid expansion of massive multilingual corpora, Multilingual Information Retrieval (MLIR) has emerged as a critical technology for global information access. MLIR enables users to retrieve semantically relevant documents from multilingual text collections using a single-language query. However, recent multilingual dense retrieval models often exhibit a strong preference for documents in the same language as the query. This leads to severe language bias, where top-ranked results are dominated by documents of specific languages, even when documents in other languages contain more semantically relevant information. To address this issue, we propose Shift, a training-free method applicable in the indexing stage. Specifically, Shift utilizes parallel translation pairs to estimate a relative language vector for each target language with respect to a source language. Subsequently, Shift corrects the language-specific offset by subtracting this relative language vector from document embeddings during indexing. Our comprehensive evaluation across four MLIR benchmarks and diverse dense retrieval models confirms that Shift can effectively mitigate language bias and enhance MLIR performance."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="jang-etal-2026-shift">
<titleInfo>
<title>SHIFT: Semantic Harmonization via Index-side Feature Transformation for Multilingual Information Retrieval</title>
</titleInfo>
<name type="personal">
<namePart type="given">Youngjoon</namePart>
<namePart type="family">Jang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Seongtae</namePart>
<namePart type="family">Hong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hyeonseok</namePart>
<namePart type="family">Moon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Heuiseok</namePart>
<namePart type="family">Lim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>With the rapid expansion of massive multilingual corpora, Multilingual Information Retrieval (MLIR) has emerged as a critical technology for global information access. MLIR enables users to retrieve semantically relevant documents from multilingual text collections using a single-language query. However, recent multilingual dense retrieval models often exhibit a strong preference for documents in the same language as the query. This leads to severe language bias, where top-ranked results are dominated by documents of specific languages, even when documents in other languages contain more semantically relevant information. To address this issue, we propose Shift, a training-free method applicable in the indexing stage. Specifically, Shift utilizes parallel translation pairs to estimate a relative language vector for each target language with respect to a source language. Subsequently, Shift corrects the language-specific offset by subtracting this relative language vector from document embeddings during indexing. Our comprehensive evaluation across four MLIR benchmarks and diverse dense retrieval models confirms that Shift can effectively mitigate language bias and enhance MLIR performance.</abstract>
<identifier type="citekey">jang-etal-2026-shift</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.109/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>2247</start>
<end>2269</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SHIFT: Semantic Harmonization via Index-side Feature Transformation for Multilingual Information Retrieval
%A Jang, Youngjoon
%A Hong, Seongtae
%A Moon, Hyeonseok
%A Lim, Heuiseok
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F jang-etal-2026-shift
%X With the rapid expansion of massive multilingual corpora, Multilingual Information Retrieval (MLIR) has emerged as a critical technology for global information access. MLIR enables users to retrieve semantically relevant documents from multilingual text collections using a single-language query. However, recent multilingual dense retrieval models often exhibit a strong preference for documents in the same language as the query. This leads to severe language bias, where top-ranked results are dominated by documents of specific languages, even when documents in other languages contain more semantically relevant information. To address this issue, we propose Shift, a training-free method applicable in the indexing stage. Specifically, Shift utilizes parallel translation pairs to estimate a relative language vector for each target language with respect to a source language. Subsequently, Shift corrects the language-specific offset by subtracting this relative language vector from document embeddings during indexing. Our comprehensive evaluation across four MLIR benchmarks and diverse dense retrieval models confirms that Shift can effectively mitigate language bias and enhance MLIR performance.
%U https://aclanthology.org/2026.findings-emnlp.109/
%P 2247-2269
Markdown (Informal)
[SHIFT: Semantic Harmonization via Index-side Feature Transformation for Multilingual Information Retrieval](https://aclanthology.org/2026.findings-emnlp.109/) (Jang et al., Findings 2026)
ACL