<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kojima-etal-2024-multilingual">
    <titleInfo>
        <title>On the Multilingual Ability of Decoder-based Pre-trained Language Models: Finding and Controlling Language-Specific Neurons</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Takeshi</namePart>
        <namePart type="family">Kojima</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Itsuki</namePart>
        <namePart type="family">Okimura</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yusuke</namePart>
        <namePart type="family">Iwasawa</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hitomi</namePart>
        <namePart type="family">Yanaka</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yutaka</namePart>
        <namePart type="family">Matsuo</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2024-06</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Kevin</namePart>
            <namePart type="family">Duh</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Helena</namePart>
            <namePart type="family">Gomez</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Steven</namePart>
            <namePart type="family">Bethard</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Mexico City, Mexico</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Current decoder-based pre-trained language models (PLMs) successfully demonstrate multilingual capabilities. However, it is unclear how these models handle multilingualism.We analyze the neuron-level internal behavior of multilingual decoder-based PLMs, Specifically examining the existence of neurons that fire “uniquely for each language” within decoder-only multilingual PLMs.We analyze six languages: English, German, French, Spanish, Chinese, and Japanese, and show that language-specific neurons are unique, with a slight overlap (\textless 5%) between languages. These neurons are mainly distributed in the models’ first and last few layers. This trend remains consistent across languages and models.Additionally, we tamper with less than 1% of the total neurons in each model during inference and demonstrate that tampering with a few language-specific neurons drastically changes the probability of target language occurrence in text generation.</abstract>
    <identifier type="citekey">kojima-etal-2024-multilingual</identifier>
    <location>
        <url>https://aclanthology.org/2024.naacl-long.384</url>
    </location>
    <part>
        <date>2024-06</date>
        <extent unit="page">
            <start>6919</start>
            <end>6971</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T On the Multilingual Ability of Decoder-based Pre-trained Language Models: Finding and Controlling Language-Specific Neurons
%A Kojima, Takeshi
%A Okimura, Itsuki
%A Iwasawa, Yusuke
%A Yanaka, Hitomi
%A Matsuo, Yutaka
%Y Duh, Kevin
%Y Gomez, Helena
%Y Bethard, Steven
%S Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
%D 2024
%8 June
%I Association for Computational Linguistics
%C Mexico City, Mexico
%F kojima-etal-2024-multilingual
%X Current decoder-based pre-trained language models (PLMs) successfully demonstrate multilingual capabilities. However, it is unclear how these models handle multilingualism.We analyze the neuron-level internal behavior of multilingual decoder-based PLMs, Specifically examining the existence of neurons that fire “uniquely for each language” within decoder-only multilingual PLMs.We analyze six languages: English, German, French, Spanish, Chinese, and Japanese, and show that language-specific neurons are unique, with a slight overlap (\textless 5%) between languages. These neurons are mainly distributed in the models’ first and last few layers. This trend remains consistent across languages and models.Additionally, we tamper with less than 1% of the total neurons in each model during inference and demonstrate that tampering with a few language-specific neurons drastically changes the probability of target language occurrence in text generation.
%U https://aclanthology.org/2024.naacl-long.384
%P 6919-6971

Markdown (Informal)

[On the Multilingual Ability of Decoder-based Pre-trained Language Models: Finding and Controlling Language-Specific Neurons](https://aclanthology.org/2024.naacl-long.384) (Kojima et al., NAACL 2024)

• On the Multilingual Ability of Decoder-based Pre-trained Language Models: Finding and Controlling Language-Specific Neurons (Kojima et al., NAACL 2024)

ACL

• Takeshi Kojima, Itsuki Okimura, Yusuke Iwasawa, Hitomi Yanaka, and Yutaka Matsuo. 2024. On the Multilingual Ability of Decoder-based Pre-trained Language Models: Finding and Controlling Language-Specific Neurons. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 6919–6971, Mexico City, Mexico. Association for Computational Linguistics.