@inproceedings{ma-etal-2025-phonothink,
    title = "{P}hono{T}hink: Improving Large Language Models' Reasoning on {C}hinese Phonological Ambiguities",
    author = "Ma, Jianfei  and
      Feng, Zhaoxin  and
      Chersoni, Emmanuele  and
      Song, Huacheng  and
      Zhang, Ziqi",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.emnlp-main.961/",
    pages = "19018--19033",
    ISBN = "979-8-89176-332-6",
    abstract = "Effectively resolving phonological ambiguities is crucial for robust natural language processing, as these ambiguities are pervasive in tasks ranging from speech-to-text, spelling correction, to offensive language detection. However, current Large Language Models (LLMs) frequently struggle to resolve such ambiguities.To address this challenge, we present a framework to enhances LLMs' phonological capability through a multiple-stage training approach. Our method begins with supervised fine-tuning on well-constructed datasets, including three subtask datasets designed to enhance the model{'}s foundational phonological knowledge, along with a synthetic dataset of step-by-step reasoning chains. Following this, we apply reinforcement learning to incentivize and stabilize its reasoning.Results show that our framework enables the base model to achieve relatively comparable performance to a much larger model. Our ablation studies reveal that subtask datasets and the synthetic dataset can simultaneously impact as complementary modular enhancers to strengthen LLMs' integrated application."
}