@inproceedings{tang-etal-2024-language,
title = "Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models",
author = "Tang, Tianyi and
Luo, Wenyang and
Huang, Haoyang and
Zhang, Dongdong and
Wang, Xiaolei and
Zhao, Xin and
Wei, Furu and
Wen, Ji-Rong",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.acl-long.309",
doi = "10.18653/v1/2024.acl-long.309",
pages = "5701--5715",
abstract = "Large language models (LLMs) demonstrate remarkable multilingual capabilities without being pre-trained on specially curated multilingual parallel corpora.It remains a challenging problem to explain the underlying mechanisms by which LLMs process multilingual texts.In this paper, we delve into the composition of Transformer architectures in LLMs to pinpoint language-specific regions.Specially, we propose a novel detection method, language activation probability entropy (LAPE), to identify language-specific neurons within LLMs.Based on LAPE, we conduct comprehensive experiments on several representative LLMs, such as LLaMA-2, BLOOM, and Mistral. Our findings indicate that LLMs{'} proficiency in processing a particular language is predominantly due to a small subset of neurons, primarily situated in the models{'} top and bottom layers.Furthermore, we showcase the feasibility to {``}steer{''} the output language of LLMs by selectively activating or deactivating language-specific neurons. Our research provides important evidence to the understanding and exploration of the multilingual capabilities of LLMs.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tang-etal-2024-language">
<titleInfo>
<title>Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tianyi</namePart>
<namePart type="family">Tang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wenyang</namePart>
<namePart type="family">Luo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haoyang</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dongdong</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaolei</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Furu</namePart>
<namePart type="family">Wei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ji-Rong</namePart>
<namePart type="family">Wen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Srikumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Large language models (LLMs) demonstrate remarkable multilingual capabilities without being pre-trained on specially curated multilingual parallel corpora.It remains a challenging problem to explain the underlying mechanisms by which LLMs process multilingual texts.In this paper, we delve into the composition of Transformer architectures in LLMs to pinpoint language-specific regions.Specially, we propose a novel detection method, language activation probability entropy (LAPE), to identify language-specific neurons within LLMs.Based on LAPE, we conduct comprehensive experiments on several representative LLMs, such as LLaMA-2, BLOOM, and Mistral. Our findings indicate that LLMs’ proficiency in processing a particular language is predominantly due to a small subset of neurons, primarily situated in the models’ top and bottom layers.Furthermore, we showcase the feasibility to “steer” the output language of LLMs by selectively activating or deactivating language-specific neurons. Our research provides important evidence to the understanding and exploration of the multilingual capabilities of LLMs.</abstract>
<identifier type="citekey">tang-etal-2024-language</identifier>
<identifier type="doi">10.18653/v1/2024.acl-long.309</identifier>
<location>
<url>https://aclanthology.org/2024.acl-long.309</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>5701</start>
<end>5715</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models
%A Tang, Tianyi
%A Luo, Wenyang
%A Huang, Haoyang
%A Zhang, Dongdong
%A Wang, Xiaolei
%A Zhao, Xin
%A Wei, Furu
%A Wen, Ji-Rong
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F tang-etal-2024-language
%X Large language models (LLMs) demonstrate remarkable multilingual capabilities without being pre-trained on specially curated multilingual parallel corpora.It remains a challenging problem to explain the underlying mechanisms by which LLMs process multilingual texts.In this paper, we delve into the composition of Transformer architectures in LLMs to pinpoint language-specific regions.Specially, we propose a novel detection method, language activation probability entropy (LAPE), to identify language-specific neurons within LLMs.Based on LAPE, we conduct comprehensive experiments on several representative LLMs, such as LLaMA-2, BLOOM, and Mistral. Our findings indicate that LLMs’ proficiency in processing a particular language is predominantly due to a small subset of neurons, primarily situated in the models’ top and bottom layers.Furthermore, we showcase the feasibility to “steer” the output language of LLMs by selectively activating or deactivating language-specific neurons. Our research provides important evidence to the understanding and exploration of the multilingual capabilities of LLMs.
%R 10.18653/v1/2024.acl-long.309
%U https://aclanthology.org/2024.acl-long.309
%U https://doi.org/10.18653/v1/2024.acl-long.309
%P 5701-5715
Markdown (Informal)
[Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models](https://aclanthology.org/2024.acl-long.309) (Tang et al., ACL 2024)
ACL
- Tianyi Tang, Wenyang Luo, Haoyang Huang, Dongdong Zhang, Xiaolei Wang, Xin Zhao, Furu Wei, and Ji-Rong Wen. 2024. Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 5701–5715, Bangkok, Thailand. Association for Computational Linguistics.