<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="a-etal-2026-ltrc">
    <titleInfo>
        <title>LTRC_Plural: Better Data and Better Rewards for Low-Resource Indic MT</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Vishnuraj</namePart>
        <namePart type="family">A</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yuvrajsinh</namePart>
        <namePart type="family">Bodana</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Heli</namePart>
        <namePart type="family">Hingrajiya</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Priyanka</namePart>
        <namePart type="family">Dasari</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Parameswari</namePart>
        <namePart type="family">Krishnamurthy</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>Proceedings of the Eleventh Conference on Machine Translation</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Barry</namePart>
            <namePart type="family">Haddow</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tom</namePart>
            <namePart type="family">Kocmi</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Philipp</namePart>
            <namePart type="family">Koehn</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Christof</namePart>
            <namePart type="family">Monz</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">979-8-89176-474-3</identifier>
    </relatedItem>
    <abstract>We present our submission to the WMT 2026 Shared Task on Low-Resource Indic Language Translation, covering four language pairs (English-Assamese, English-Mizo, English-Khasi, and English-Tagin in both directions) spanning a wide range of resource levels, from moderately-resourced Assamese down to Tagin, which barely appears in existing multilingual MT systems at all. We compare three models (NLLB-200-1.3B, Sarvam-Translate, and TranslateGemma) across three setups: fine-tuning on the official shared-task data alone, fine-tuning with added external parallel data, and a reinforcement-learning stage (RLOO) that optimizes a combined BLEU/chrF++ reward on top of the fine-tuned model. Adding external parallel data helps most for Khasi, improving English-to-Khasi by 18 BLEU, and for Mizo, where our system places first among contrastive submissions in both directions; the same augmentation leaves Assamese essentially unchanged despite contributing twice as many sentence pairs, suggesting that data provenance matters more than volume at these scales. The RLOO system degrades performance on Assamese and Mizo, where supervised fine-tuning had already produced competitive systems, but improves over the primary system on both Khasi directions and on every metric for English-to-Tagin, where it obtains the highest METEOR score of any contrastive submission. Reward-based training therefore appears most useful precisely where supervised fine-tuning has least to work with.</abstract>
    <identifier type="citekey">a-etal-2026-ltrc</identifier>
    <location>
        <url>https://aclanthology.org/2026.wmt-main.145/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>2067</start>
            <end>2077</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T LTRC_Plural: Better Data and Better Rewards for Low-Resource Indic MT
%A A, Vishnuraj
%A Bodana, Yuvrajsinh
%A Hingrajiya, Heli
%A Dasari, Priyanka
%A Krishnamurthy, Parameswari
%Y Haddow, Barry
%Y Kocmi, Tom
%Y Koehn, Philipp
%Y Monz, Christof
%S Proceedings of the Eleventh Conference on Machine Translation
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-474-3
%F a-etal-2026-ltrc
%X We present our submission to the WMT 2026 Shared Task on Low-Resource Indic Language Translation, covering four language pairs (English-Assamese, English-Mizo, English-Khasi, and English-Tagin in both directions) spanning a wide range of resource levels, from moderately-resourced Assamese down to Tagin, which barely appears in existing multilingual MT systems at all. We compare three models (NLLB-200-1.3B, Sarvam-Translate, and TranslateGemma) across three setups: fine-tuning on the official shared-task data alone, fine-tuning with added external parallel data, and a reinforcement-learning stage (RLOO) that optimizes a combined BLEU/chrF++ reward on top of the fine-tuned model. Adding external parallel data helps most for Khasi, improving English-to-Khasi by 18 BLEU, and for Mizo, where our system places first among contrastive submissions in both directions; the same augmentation leaves Assamese essentially unchanged despite contributing twice as many sentence pairs, suggesting that data provenance matters more than volume at these scales. The RLOO system degrades performance on Assamese and Mizo, where supervised fine-tuning had already produced competitive systems, but improves over the primary system on both Khasi directions and on every metric for English-to-Tagin, where it obtains the highest METEOR score of any contrastive submission. Reward-based training therefore appears most useful precisely where supervised fine-tuning has least to work with.
%U https://aclanthology.org/2026.wmt-main.145/
%P 2067-2077

Markdown (Informal)

[LTRC_Plural: Better Data and Better Rewards for Low-Resource Indic MT](https://aclanthology.org/2026.wmt-main.145/) (A et al., WMT 2026)

• LTRC_Plural: Better Data and Better Rewards for Low-Resource Indic MT (A et al., WMT 2026)

ACL

• Vishnuraj A, Yuvrajsinh Bodana, Heli Hingrajiya, Priyanka Dasari, and Parameswari Krishnamurthy. 2026. LTRC_Plural: Better Data and Better Rewards for Low-Resource Indic MT. In Proceedings of the Eleventh Conference on Machine Translation, pages 2067–2077, Budapest, Hungary. Association for Computational Linguistics.