<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="xie-etal-2025-language">
    <titleInfo>
        <title>Can Language Neuron Intervention Reduce Non-Target Language Output?</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Suchun</namePart>
        <namePart type="family">Xie</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hwichan</namePart>
        <namePart type="family">Kim</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Shota</namePart>
        <namePart type="family">Sasaki</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Kosuke</namePart>
        <namePart type="family">Yamada</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jun</namePart>
        <namePart type="family">Suzuki</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 8th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Yonatan</namePart>
            <namePart type="family">Belinkov</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Aaron</namePart>
            <namePart type="family">Mueller</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Najoung</namePart>
            <namePart type="family">Kim</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Hosein</namePart>
            <namePart type="family">Mohebbi</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Hanjie</namePart>
            <namePart type="family">Chen</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Dana</namePart>
            <namePart type="family">Arad</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Gabriele</namePart>
            <namePart type="family">Sarti</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Suzhou, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-346-3</identifier>
    </relatedItem>
    <abstract>Large language models (LLMs) often fail to generate text in the intended target language, particularly in non-English interactions. Concurrently, recent work has explored Language Neuron Intervention (LNI) as a promising technique for steering output language. In this paper, we re-evaluate LNI in more practical scenarios—specifically with instruction-tuned models and prompts that explicitly specify the target language. Our experiments show that while LNI also shows potential in such practical scenarios, its average effect is limited and unstable across models and tasks, with a 0.83% reduction in undesired language output and a 0.1% improvement in performance. Our further analysis identifies two key factors for LNI’s limitation: (1) LNI affects both the output language and the content semantics, making it hard to control one without affecting the other, which explains the weak performance gains. (2) LNI increases the target language token probabilities, but they often remain below the top-1generation threshold, resulting in failure to generate the target language in most cases. Our results highlight both the potential and limitations of LNI, paving the way for future improvements.</abstract>
    <identifier type="citekey">xie-etal-2025-language</identifier>
    <identifier type="doi">10.18653/v1/2025.blackboxnlp-1.26</identifier>
    <location>
        <url>https://aclanthology.org/2025.blackboxnlp-1.26/</url>
    </location>
    <part>
        <date>2025-11</date>
        <extent unit="page">
            <start>452</start>
            <end>466</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Can Language Neuron Intervention Reduce Non-Target Language Output?
%A Xie, Suchun
%A Kim, Hwichan
%A Sasaki, Shota
%A Yamada, Kosuke
%A Suzuki, Jun
%Y Belinkov, Yonatan
%Y Mueller, Aaron
%Y Kim, Najoung
%Y Mohebbi, Hosein
%Y Chen, Hanjie
%Y Arad, Dana
%Y Sarti, Gabriele
%S Proceedings of the 8th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-346-3
%F xie-etal-2025-language
%X Large language models (LLMs) often fail to generate text in the intended target language, particularly in non-English interactions. Concurrently, recent work has explored Language Neuron Intervention (LNI) as a promising technique for steering output language. In this paper, we re-evaluate LNI in more practical scenarios—specifically with instruction-tuned models and prompts that explicitly specify the target language. Our experiments show that while LNI also shows potential in such practical scenarios, its average effect is limited and unstable across models and tasks, with a 0.83% reduction in undesired language output and a 0.1% improvement in performance. Our further analysis identifies two key factors for LNI’s limitation: (1) LNI affects both the output language and the content semantics, making it hard to control one without affecting the other, which explains the weak performance gains. (2) LNI increases the target language token probabilities, but they often remain below the top-1generation threshold, resulting in failure to generate the target language in most cases. Our results highlight both the potential and limitations of LNI, paving the way for future improvements.
%R 10.18653/v1/2025.blackboxnlp-1.26
%U https://aclanthology.org/2025.blackboxnlp-1.26/
%U https://doi.org/10.18653/v1/2025.blackboxnlp-1.26
%P 452-466

Markdown (Informal)

[Can Language Neuron Intervention Reduce Non-Target Language Output?](https://aclanthology.org/2025.blackboxnlp-1.26/) (Xie et al., BlackboxNLP 2025)

• Can Language Neuron Intervention Reduce Non-Target Language Output? (Xie et al., BlackboxNLP 2025)

ACL

• Suchun Xie, Hwichan Kim, Shota Sasaki, Kosuke Yamada, and Jun Suzuki. 2025. Can Language Neuron Intervention Reduce Non-Target Language Output?. In Proceedings of the 8th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP, pages 452–466, Suzhou, China. Association for Computational Linguistics.