@inproceedings{minh-etal-2025-challenge,
    title = "Challenge Track: {JHARNA}-{MT}: A Copy-Augmented Hybrid of {L}o{RA}-Tuned {NLLB} and Lexical {SMT} with Minimum {B}ayes Risk Decoding for Low-Resource {I}ndic Languages",
    author = "Minh, Dao Sy Duy  and
      Huynh, Trung Kiet  and
      Nguyen, Tran Chi  and
      Lam, Phu Quy Nguyen  and
      Pham, Phu-Hoa  and
      Dương, Nguyễn {\DJ}{\`i}nh H{\`a}  and
      Dinh, Dien  and
      Nguyen, Long HB",
    editor = "Shukla, Ankita  and
      Kumar, Sandeep  and
      Bedi, Amrit Singh  and
      Chakraborty, Tanmoy",
    booktitle = "Proceedings of the 1st Workshop on Multimodal Models for Low-Resource Contexts and Social Impact (MMLoSo 2025)",
    month = dec,
    year = "2025",
    address = "Mumbai, India",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/master-new-author-system-docs/2025.mmloso-1.13/",
    pages = "114--120",
    ISBN = "979-8-89176-311-1",
    abstract = "This paper describes JHARNA-MT, our system for the MMLoSo 2025 Shared Task on translation between high-resource languages (Hindi, English) and four low-resource Indic tribal languages: Bhili, Gondi, Mundari, and Santali. The task poses significant challenges, including data sparsity, morphological richness, and structural divergence across language pairs. To address these, we propose a hybrid translation pipeline that integrates non-parametric retrieval, lexical statistical machine translation (SMT), and LoRA-tuned NLLB-200 neural machine translation under a unified Minimum Bayes Risk (MBR) decoding framework. Exact and fuzzy retrieval exploit redundancy in government and administrative texts, SMT with diagonal alignment priors and back-translation provides lexically faithful hypotheses, and the NLLB-LoRA component contributes fluent neural candidates. MBR decoding selects consensus translations using a metric-matched utility based on a weighted combination of BLEU and chrF, mitigating the complementary error modes of SMT and NMT. Our final system, further enhanced with script-aware digit normalization and entity-preserving post-processing, achieves a private leaderboard score of 186.37 and ranks 2nd overall in the shared task, with ablation studies confirming the contribution of each component."
}