<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="rao-etal-2026-decepticons">
    <titleInfo>
        <title>Decepticons-IIITH : Morphology-Aware Lexicon Filtering of Back-Translated Data for Low-Resource English-to-Indic Translation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Soujanya</namePart>
        <namePart type="family">Rao</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Sakhil</namePart>
        <namePart type="family">Taneja</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Krishna</namePart>
        <namePart type="family">Chaitanya</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Radhika</namePart>
        <namePart type="family">Mamidi</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>This study tackles the WMT 2026 Low-Resource Indic Language Translation task, covering English → Assamese, Bodo, Mizo and Nagamese. Low-resource languages lack parallel corpora of sufficient size and quality to train machine translation models. To address this challenge, we propose Morphology-Aware Lexicon Filtering of Back-Translated data to generate good-quality parallel corpora. Initially, monolingual text from publicly available sources is back-translated to English with an NLLB-200-3.3B model fine-tuned on the official training data. To filter this synthetic data, we introduce MALM (Morphology-Aware Lexicon Matching), a filter that checks whether words from a bilingual lexicon are preserved in a translation, using Morfessor stems to handle rich morphology and edit-distance matching to handle borrowed words. For our primary method, training occurs in two stages: fine-tuning on filtered synthetic (back-translated) data followed by fine-tuning on the official parallel data. Whereas for the contrastive method, we are only fine-tuning on the official parallel data. We fine-tune the NLLB-200-3.3B model for Assamese, Mizo, Nagamese and the IndicTrans2-1B model for Bodo. Our primary systems ranked first among primary submissions on two Category-2 pairs viz. English → Bodo and English → Nagamese. On Nagamese, where only 2,000 official sentence pairs exist, of the two methods we used, the primary system improved by a BLEU score of 8.33.</abstract>
    <identifier type="citekey">rao-etal-2026-decepticons</identifier>
    <location>
        <url>https://aclanthology.org/2026.wmt-main.160/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>2207</start>
            <end>2212</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Decepticons-IIITH : Morphology-Aware Lexicon Filtering of Back-Translated Data for Low-Resource English-to-Indic Translation
%A Rao, Soujanya
%A Taneja, Sakhil
%A Chaitanya, Krishna
%A Mamidi, Radhika
%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 rao-etal-2026-decepticons
%X This study tackles the WMT 2026 Low-Resource Indic Language Translation task, covering English → Assamese, Bodo, Mizo and Nagamese. Low-resource languages lack parallel corpora of sufficient size and quality to train machine translation models. To address this challenge, we propose Morphology-Aware Lexicon Filtering of Back-Translated data to generate good-quality parallel corpora. Initially, monolingual text from publicly available sources is back-translated to English with an NLLB-200-3.3B model fine-tuned on the official training data. To filter this synthetic data, we introduce MALM (Morphology-Aware Lexicon Matching), a filter that checks whether words from a bilingual lexicon are preserved in a translation, using Morfessor stems to handle rich morphology and edit-distance matching to handle borrowed words. For our primary method, training occurs in two stages: fine-tuning on filtered synthetic (back-translated) data followed by fine-tuning on the official parallel data. Whereas for the contrastive method, we are only fine-tuning on the official parallel data. We fine-tune the NLLB-200-3.3B model for Assamese, Mizo, Nagamese and the IndicTrans2-1B model for Bodo. Our primary systems ranked first among primary submissions on two Category-2 pairs viz. English → Bodo and English → Nagamese. On Nagamese, where only 2,000 official sentence pairs exist, of the two methods we used, the primary system improved by a BLEU score of 8.33.
%U https://aclanthology.org/2026.wmt-main.160/
%P 2207-2212

Markdown (Informal)

[Decepticons-IIITH : Morphology-Aware Lexicon Filtering of Back-Translated Data for Low-Resource English-to-Indic Translation](https://aclanthology.org/2026.wmt-main.160/) (Rao et al., WMT 2026)

• Decepticons-IIITH : Morphology-Aware Lexicon Filtering of Back-Translated Data for Low-Resource English-to-Indic Translation (Rao et al., WMT 2026)

ACL

• Soujanya Rao, Sakhil Taneja, Krishna Chaitanya, and Radhika Mamidi. 2026. Decepticons-IIITH : Morphology-Aware Lexicon Filtering of Back-Translated Data for Low-Resource English-to-Indic Translation. In Proceedings of the Eleventh Conference on Machine Translation, pages 2207–2212, Budapest, Hungary. Association for Computational Linguistics.