@inproceedings{tran-etal-2026-representation,
    title = "Representation-Aware Prompting for Zero-Shot {M}arathi Text Classification: {IPA}, {R}omanization, Repetition",
    author = "Tran, Van-Hien  and
      Vu, Huy Hien  and
      Tanaka, Hideki  and
      Utiyama, Masao",
    editor = "Hettiarachchi, Hansi  and
      Ranasinghe, Tharindu  and
      Plum, Alistair  and
      Rayson, Paul  and
      Mitkov, Ruslan  and
      Gaber, Mohamed  and
      Premasiri, Damith  and
      Tan, Fiona Anting  and
      Uyangodage, Lasitha",
    booktitle = "Proceedings of the Second Workshop on Language Models for Low-Resource Languages ({L}o{R}es{LM} 2026)",
    month = mar,
    year = "2026",
    address = "Rabat, Morocco",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/manual-author-scripts/2026.loreslm-1.37/",
    pages = "436--443",
    ISBN = "979-8-89176-377-7",
    abstract = "Large language models (LLMs) often underperform in zero-shot text classification for low-resource, non-Latin languages due to script and tokenization mismatches. We propose \textit{representation-aware prompting} for Marathi that augments the original script with International Phonetic Alphabet (IPA) transcriptions, romanization, or a repetition-based fallback when external converters are unavailable. Experiments with two instruction-tuned LLMs on Marathi sentiment analysis and hate detection show consistent gains over script-only prompting (up to +2.6 accuracy points). We further find that the most effective augmentation is model-dependent, and that combining all variants is not consistently beneficial, suggesting that concise, targeted cues are preferable in zero-shot settings."
}