@inproceedings{liu-etal-2026-silp,
title = "{S}i{LP}: Enhancing Non-Dominant Language Capabilities with a Selective Bidirectional Language Projection Framework",
author = "Liu, Junpeng and
Li, Jiuyi and
Huang, Kaiyu and
Jin, Bo and
Huang, Degen and
Xiong, Hui",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.acl-long.2005/",
pages = "43307--43325",
ISBN = "979-8-89176-390-6",
abstract = "Current large language models (LLMs) often exhibit performance imbalances between dominant languages (e.g., English) and non-dominant ones due to the skewed distribution of pretraining data. A common strategy to address this issue is to enhance cross-lingual alignment, thereby facilitating non-dominant language processing. However, existing methods typically rely on additional training objectives or language-specific parameters, which increase training complexity and cost. In this work, we propose a selective bidirectional language projection framework that enables efficient multilingual alignment and language shift using the intrinsic parameters. Specifically, we first identify the layers most sensitive to language projection between non-dominant and dominant languages through neuron activation analysis. We then perform sequential language projection within the selected layers by mapping non-dominant representations into the dominant language space and reverting them before generation. The bidirectional projection benefits the subsequent instruction tuning in non-dominant languages. Experiments on seven benchmarks demonstrate that our method remarkably enhances the performance of non-dominant languages. Further analyses indicate that our method learns better internal representations and exhibits strong generalization capabilities."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2026-silp">
<titleInfo>
<title>SiLP: Enhancing Non-Dominant Language Capabilities with a Selective Bidirectional Language Projection Framework</title>
</titleInfo>
<name type="personal">
<namePart type="given">Junpeng</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiuyi</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kaiyu</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bo</namePart>
<namePart type="family">Jin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Degen</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hui</namePart>
<namePart type="family">Xiong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-390-6</identifier>
</relatedItem>
<abstract>Current large language models (LLMs) often exhibit performance imbalances between dominant languages (e.g., English) and non-dominant ones due to the skewed distribution of pretraining data. A common strategy to address this issue is to enhance cross-lingual alignment, thereby facilitating non-dominant language processing. However, existing methods typically rely on additional training objectives or language-specific parameters, which increase training complexity and cost. In this work, we propose a selective bidirectional language projection framework that enables efficient multilingual alignment and language shift using the intrinsic parameters. Specifically, we first identify the layers most sensitive to language projection between non-dominant and dominant languages through neuron activation analysis. We then perform sequential language projection within the selected layers by mapping non-dominant representations into the dominant language space and reverting them before generation. The bidirectional projection benefits the subsequent instruction tuning in non-dominant languages. Experiments on seven benchmarks demonstrate that our method remarkably enhances the performance of non-dominant languages. Further analyses indicate that our method learns better internal representations and exhibits strong generalization capabilities.</abstract>
<identifier type="citekey">liu-etal-2026-silp</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.2005/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>43307</start>
<end>43325</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SiLP: Enhancing Non-Dominant Language Capabilities with a Selective Bidirectional Language Projection Framework
%A Liu, Junpeng
%A Li, Jiuyi
%A Huang, Kaiyu
%A Jin, Bo
%A Huang, Degen
%A Xiong, Hui
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-390-6
%F liu-etal-2026-silp
%X Current large language models (LLMs) often exhibit performance imbalances between dominant languages (e.g., English) and non-dominant ones due to the skewed distribution of pretraining data. A common strategy to address this issue is to enhance cross-lingual alignment, thereby facilitating non-dominant language processing. However, existing methods typically rely on additional training objectives or language-specific parameters, which increase training complexity and cost. In this work, we propose a selective bidirectional language projection framework that enables efficient multilingual alignment and language shift using the intrinsic parameters. Specifically, we first identify the layers most sensitive to language projection between non-dominant and dominant languages through neuron activation analysis. We then perform sequential language projection within the selected layers by mapping non-dominant representations into the dominant language space and reverting them before generation. The bidirectional projection benefits the subsequent instruction tuning in non-dominant languages. Experiments on seven benchmarks demonstrate that our method remarkably enhances the performance of non-dominant languages. Further analyses indicate that our method learns better internal representations and exhibits strong generalization capabilities.
%U https://aclanthology.org/2026.acl-long.2005/
%P 43307-43325
Markdown (Informal)
[SiLP: Enhancing Non-Dominant Language Capabilities with a Selective Bidirectional Language Projection Framework](https://aclanthology.org/2026.acl-long.2005/) (Liu et al., ACL 2026)
ACL