<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mazumder-etal-2026-neither">
    <titleInfo>
        <title>Neither Here Nor There: Cross-Lingual Representation Dynamics of Code-Mixed Text — A Hindi–English Case Study</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Debajyoti</namePart>
        <namePart type="family">Mazumder</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Divyansh</namePart>
        <namePart type="family">Pathak</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Prashant</namePart>
        <namePart type="family">Kodali</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jasabanta</namePart>
        <namePart type="family">Patro</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Hung-Yi</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sunipa</namePart>
            <namePart type="family">Dev</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jessy</namePart>
            <namePart type="family">Li</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Desmond</namePart>
            <namePart type="family">Elliott</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-462-0</identifier>
    </relatedItem>
    <abstract>Multilingual encoder-based language models are widely used for code-mixed analysis, yet their internal representations of code-mixed inputs—and their relationship to the constituent languages—remain poorly understood. Using Hindi–English as a case study, we construct a unified trilingual corpus of parallel English, Hindi (Devanagari), and Romanized code-mixed sentences. We then probe cross-lingual representation alignment in standard multilingual encoders and their code-mix-adapted variants using CKA, token-level saliency, and entropy-based uncertainty analysis. We find that while standard models align English and Hindi well, code-mixed inputs remain loosely connected to either language — and that continued pre-training on code-mixed data improves English–code-mixed alignment at the cost of English–Hindi alignment. Interpretability analyses further reveal a clear asymmetry: models process code-mixed text through an English-dominant semantic subspace, while native-script Hindi provides complementary signals that reduce representational uncertainty. Motivated by these findings, we introduce a trilingual post-training alignment objective that brings code-mixed representations closer to both constituent languages simultaneously, yielding more balanced cross-lingual alignment and downstream gains on sentiment analysis and hate speech detection — showing that grounding code-mixed representations in their constituent languages meaningfully helps cross-lingual understanding. Code is available at: https://github.com/debajyotimaz/tri_align_EMNLP_2026.</abstract>
    <identifier type="citekey">mazumder-etal-2026-neither</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.1748/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>37390</start>
            <end>37415</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Neither Here Nor There: Cross-Lingual Representation Dynamics of Code-Mixed Text — A Hindi–English Case Study
%A Mazumder, Debajyoti
%A Pathak, Divyansh
%A Kodali, Prashant
%A Patro, Jasabanta
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F mazumder-etal-2026-neither
%X Multilingual encoder-based language models are widely used for code-mixed analysis, yet their internal representations of code-mixed inputs—and their relationship to the constituent languages—remain poorly understood. Using Hindi–English as a case study, we construct a unified trilingual corpus of parallel English, Hindi (Devanagari), and Romanized code-mixed sentences. We then probe cross-lingual representation alignment in standard multilingual encoders and their code-mix-adapted variants using CKA, token-level saliency, and entropy-based uncertainty analysis. We find that while standard models align English and Hindi well, code-mixed inputs remain loosely connected to either language — and that continued pre-training on code-mixed data improves English–code-mixed alignment at the cost of English–Hindi alignment. Interpretability analyses further reveal a clear asymmetry: models process code-mixed text through an English-dominant semantic subspace, while native-script Hindi provides complementary signals that reduce representational uncertainty. Motivated by these findings, we introduce a trilingual post-training alignment objective that brings code-mixed representations closer to both constituent languages simultaneously, yielding more balanced cross-lingual alignment and downstream gains on sentiment analysis and hate speech detection — showing that grounding code-mixed representations in their constituent languages meaningfully helps cross-lingual understanding. Code is available at: https://github.com/debajyotimaz/tri_align_EMNLP_2026.
%U https://aclanthology.org/2026.findings-emnlp.1748/
%P 37390-37415

Markdown (Informal)

[Neither Here Nor There: Cross-Lingual Representation Dynamics of Code-Mixed Text — A Hindi–English Case Study](https://aclanthology.org/2026.findings-emnlp.1748/) (Mazumder et al., Findings 2026)

• Neither Here Nor There: Cross-Lingual Representation Dynamics of Code-Mixed Text — A Hindi–English Case Study (Mazumder et al., Findings 2026)

ACL

• Debajyoti Mazumder, Divyansh Pathak, Prashant Kodali, and Jasabanta Patro. 2026. Neither Here Nor There: Cross-Lingual Representation Dynamics of Code-Mixed Text — A Hindi–English Case Study. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 37390–37415, Budapest, Hungary. Association for Computational Linguistics.