@inproceedings{anilkumar-etal-2026-limp,
    title = "{LIMP}: Linguistically-Informed Multi-Strategy Prompting for {T}elugu Multi-Turn Dialogue Generation",
    author = "Anilkumar, Arjungopal  and
      Menon, Suryansh Ram  and
      S, Divagar  and
      B, Premjith",
    editor = "Chakravarthi, Bharathi Raja  and
      Priyadharshini, Ruba  and
      Madasamy, Anand Kumar  and
      Thavareesan, Sajeetha  and
      Rajiakodi, Saranya  and
      Navaneethakrishnan, Subalalitha  and
      Chinnappa, Dhivya  and
      Palani, Balasubramanian  and
      Subramanian, Malliga  and
      Shanmugavadivel, Kogilavani  and
      Rajalakshmi, Ratnavel",
    booktitle = "Proceedings of the Sixth Workshop on Speech, Vision, and Language Technologies for {D}ravidian Languages",
    month = jul,
    year = "2026",
    address = "Underline (Virtual)",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl-workshops/2026.dravidianlangtech-1.5/",
    pages = "32--41",
    ISBN = "979-8-89176-401-9",
    abstract = "Generating contextually coherent multi-turn dialogue in Telugu requires resolving three deeply interacting constraints absent from generic LLM prompting: morphologically encoded social hierarchy (honorific verb conjugations), strict SOV agglutinative syntax, and culturally governed emotional logic formalised in Natyashastra rasa theory (Bharata Muni, 1951). We introduce LIMP (Linguistically-Informed Multi-Strategy Prompting), an inference-time, training-free framework that injects expert linguistic and cultural knowledge into prompt structure, requiring no fine-tuning or labelled data. We empirically evaluate two strategies on 10,000 stratified evaluation instances from the IndicDialogue Telugu corpus (Arnob et al., 2024): LIMP-RAW, a dense constraint prompt, and LIMP-COT, a six-stage analytical scaffold grounded in rasa theory and Telugu morphological grammar. Our primary finding is that LIMP-COT achieves approximately 2{\texttimes} higher morphosyntactic surface fidelity than LIMP-RAW on GEMMA-3-1B-IT (Gemma Team, Google DeepMind, 2025) (1B parameters): Jaccard = 0.0436 vs. 0.0211, Dice = 0.0792 vs. 0.0411 (p {\ensuremath{<}} 0.001, Cohen{'}s d = 0.57), demonstrating that sequential analytical commitment to linguistic constraints produces more form-faithful Telugu than holistic constraint injection. Concurrently, LIMP-RAW achieves near-ceiling semantic fidelity (BERTSCORE F1 = 0.9709), exceeding both LIMP-COT (0.9637) and SARVAM-1 (Sarvam AI, 2024) (2B, Indic-pretrained; 0.9680) on this dimension. This semantic{--}lexical dissociation{---}no single configuration dominates across both metric classes{---}is itself a substantive finding: in agglutinative Telugu, semantic paraphrase fidelity and morphosyntactic surface fidelity are orthogonal evaluation dimensions. On lexical metrics specifically, LIMP-COT with a 1B general-purpose model surpasses SARVAM-1 under matched prompting (Jaccard = 0.0436 vs. 0.0052), suggesting that structured linguistic scaffolding is a stronger lever than parametric scale for form-faithful generation."
}