<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kanojia-2025-challenge">
    <titleInfo>
        <title>Challenge Track: Divide and Translate: Parameter Isolation with Encoder Freezing for Low-Resource Indic NMT</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Vaibhav</namePart>
        <namePart type="family">Kanojia</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-12</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 1st Workshop on Multimodal Models for Low-Resource Contexts and Social Impact (MMLoSo 2025)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Ankita</namePart>
            <namePart type="family">Shukla</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sandeep</namePart>
            <namePart type="family">Kumar</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Amrit</namePart>
            <namePart type="given">Singh</namePart>
            <namePart type="family">Bedi</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tanmoy</namePart>
            <namePart type="family">Chakraborty</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Mumbai, India</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-311-1</identifier>
    </relatedItem>
    <abstract>We present a direction-specialized neural machine translation framework for ultra-low-resource Indic and tribal languages, including Bhili, Gondi, Mundari, and Santali. Using the NLLB-600M backbone, we freeze the multilingual encoder and fine-tune direction-specific decoders to reduce negative transfer and improve morphological fidelity under severe data scarcity. Our system is trained with leakage-safe splits, bitext reversal augmentation, and memory-efficient mixed-precision optimization. On the official MMLoSo 2025 Kaggle benchmark, we achieve a public score of 171.4 and a private score of 161.1, demonstrating stable generalization in highly noisy low-resource conditions.</abstract>
    <identifier type="citekey">kanojia-2025-challenge</identifier>
    <location>
        <url>https://aclanthology.org/2025.mmloso-1.12/</url>
    </location>
    <part>
        <date>2025-12</date>
        <extent unit="page">
            <start>109</start>
            <end>113</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Challenge Track: Divide and Translate: Parameter Isolation with Encoder Freezing for Low-Resource Indic NMT
%A Kanojia, Vaibhav
%Y Shukla, Ankita
%Y Kumar, Sandeep
%Y Bedi, Amrit Singh
%Y Chakraborty, Tanmoy
%S Proceedings of the 1st Workshop on Multimodal Models for Low-Resource Contexts and Social Impact (MMLoSo 2025)
%D 2025
%8 December
%I Association for Computational Linguistics
%C Mumbai, India
%@ 979-8-89176-311-1
%F kanojia-2025-challenge
%X We present a direction-specialized neural machine translation framework for ultra-low-resource Indic and tribal languages, including Bhili, Gondi, Mundari, and Santali. Using the NLLB-600M backbone, we freeze the multilingual encoder and fine-tune direction-specific decoders to reduce negative transfer and improve morphological fidelity under severe data scarcity. Our system is trained with leakage-safe splits, bitext reversal augmentation, and memory-efficient mixed-precision optimization. On the official MMLoSo 2025 Kaggle benchmark, we achieve a public score of 171.4 and a private score of 161.1, demonstrating stable generalization in highly noisy low-resource conditions.
%U https://aclanthology.org/2025.mmloso-1.12/
%P 109-113

Markdown (Informal)

[Challenge Track: Divide and Translate: Parameter Isolation with Encoder Freezing for Low-Resource Indic NMT](https://aclanthology.org/2025.mmloso-1.12/) (Kanojia, MMLoSo 2025)

• Challenge Track: Divide and Translate: Parameter Isolation with Encoder Freezing for Low-Resource Indic NMT (Kanojia, MMLoSo 2025)

ACL

• Vaibhav Kanojia. 2025. Challenge Track: Divide and Translate: Parameter Isolation with Encoder Freezing for Low-Resource Indic NMT. In Proceedings of the 1st Workshop on Multimodal Models for Low-Resource Contexts and Social Impact (MMLoSo 2025), pages 109–113, Mumbai, India. Association for Computational Linguistics.