@inproceedings{wang-etal-2025-class,
    title = "Class Distillation with Mahalanobis Contrast: An Efficient Training Paradigm for Pragmatic Language Understanding Tasks",
    author = "Wang, Chenlu  and
      Lyu, Weimin  and
      Banerjee, Ritwik",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/landing_page/2025.acl-long.1424/",
    pages = "29428--29442",
    ISBN = "979-8-89176-251-0",
    abstract = "Detecting deviant language such as sexism, or nuanced language such as metaphors or sarcasm, is crucial for enhancing the safety, clarity, and interpretation of social interactions. While existing classifiers deliver strong results on these tasks, they often come with significant computational cost and high data demands. In this work, we propose \textbf{Cla}ss \textbf{D}istillation (ClaD), a novel training paradigm that targets the core challenge: distilling a small, well-defined target class from a highly diverse and heterogeneous background. ClaD integrates two key innovations: (i) a loss function informed by the structural properties of class distributions, based on Mahalanobis distance, and (ii) an interpretable decision algorithm optimized for class separation. Across three benchmark detection tasks {--} sexism, metaphor, and sarcasm {--} ClaD outperforms competitive baselines, and even with smaller language models and orders of magnitude fewer parameters, achieves performance comparable to several large language models. These results demonstrate ClaD as an efficient tool for pragmatic language understanding tasks that require gleaning a small target class from a larger heterogeneous background."
}