@inproceedings{wang-etal-2025-calm,
title = "{CALM}: Unleashing the Cross-Lingual Self-Aligning Ability of Language Model Question Answering",
author = "Wang, Yumeng and
Fan, Zhiyuan and
Wang, Qingyun and
Fung, Yi R. and
Ji, Heng",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Findings of the Association for Computational Linguistics: NAACL 2025",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-naacl.152/",
doi = "10.18653/v1/2025.findings-naacl.152",
pages = "2809--2817",
ISBN = "979-8-89176-195-7",
abstract = "Large Language Models (LLMs) are pretrained on extensive multilingual corpora to acquire both language-specific cultural knowledge and general knowledge. Ideally, while LLMs should provide consistent responses to culture-independent questions across languages, we observe significant performance disparities. To address this, we explore the **C**ross-Lingual Self-**A**ligning ability of **L**anguage **M**odels (**CALM**) to align knowledge across languages. Specifically, for a given question, we sample multiple responses across different languages and select the most self-consistent response as the target, leaving the remaining responses as negative examples. We then employ direct preference optimization (DPO) to align the model{'}s knowledge across different languages. Evaluations on the MEDQA and X-CSQA datasets demonstrate CALM{'}s effectiveness in enhancing cross-lingual knowledge question answering, both in zero-shot and retrieval-augmented settings. We also found that increasing the number of languages involved in CALM training leads to higher accuracy and consistency. We offer a qualitative analysis of how cross-lingual consistency can enhance knowledge alignment and explore the method{'}s generalizability."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2025-calm">
<titleInfo>
<title>CALM: Unleashing the Cross-Lingual Self-Aligning Ability of Language Model Question Answering</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yumeng</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhiyuan</namePart>
<namePart type="family">Fan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qingyun</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="given">R</namePart>
<namePart type="family">Fung</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Heng</namePart>
<namePart type="family">Ji</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: NAACL 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="family">Chiruzzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-195-7</identifier>
</relatedItem>
<abstract>Large Language Models (LLMs) are pretrained on extensive multilingual corpora to acquire both language-specific cultural knowledge and general knowledge. Ideally, while LLMs should provide consistent responses to culture-independent questions across languages, we observe significant performance disparities. To address this, we explore the **C**ross-Lingual Self-**A**ligning ability of **L**anguage **M**odels (**CALM**) to align knowledge across languages. Specifically, for a given question, we sample multiple responses across different languages and select the most self-consistent response as the target, leaving the remaining responses as negative examples. We then employ direct preference optimization (DPO) to align the model’s knowledge across different languages. Evaluations on the MEDQA and X-CSQA datasets demonstrate CALM’s effectiveness in enhancing cross-lingual knowledge question answering, both in zero-shot and retrieval-augmented settings. We also found that increasing the number of languages involved in CALM training leads to higher accuracy and consistency. We offer a qualitative analysis of how cross-lingual consistency can enhance knowledge alignment and explore the method’s generalizability.</abstract>
<identifier type="citekey">wang-etal-2025-calm</identifier>
<identifier type="doi">10.18653/v1/2025.findings-naacl.152</identifier>
<location>
<url>https://aclanthology.org/2025.findings-naacl.152/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>2809</start>
<end>2817</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CALM: Unleashing the Cross-Lingual Self-Aligning Ability of Language Model Question Answering
%A Wang, Yumeng
%A Fan, Zhiyuan
%A Wang, Qingyun
%A Fung, Yi R.
%A Ji, Heng
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Findings of the Association for Computational Linguistics: NAACL 2025
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-195-7
%F wang-etal-2025-calm
%X Large Language Models (LLMs) are pretrained on extensive multilingual corpora to acquire both language-specific cultural knowledge and general knowledge. Ideally, while LLMs should provide consistent responses to culture-independent questions across languages, we observe significant performance disparities. To address this, we explore the **C**ross-Lingual Self-**A**ligning ability of **L**anguage **M**odels (**CALM**) to align knowledge across languages. Specifically, for a given question, we sample multiple responses across different languages and select the most self-consistent response as the target, leaving the remaining responses as negative examples. We then employ direct preference optimization (DPO) to align the model’s knowledge across different languages. Evaluations on the MEDQA and X-CSQA datasets demonstrate CALM’s effectiveness in enhancing cross-lingual knowledge question answering, both in zero-shot and retrieval-augmented settings. We also found that increasing the number of languages involved in CALM training leads to higher accuracy and consistency. We offer a qualitative analysis of how cross-lingual consistency can enhance knowledge alignment and explore the method’s generalizability.
%R 10.18653/v1/2025.findings-naacl.152
%U https://aclanthology.org/2025.findings-naacl.152/
%U https://doi.org/10.18653/v1/2025.findings-naacl.152
%P 2809-2817
Markdown (Informal)
[CALM: Unleashing the Cross-Lingual Self-Aligning Ability of Language Model Question Answering](https://aclanthology.org/2025.findings-naacl.152/) (Wang et al., Findings 2025)
ACL