@inproceedings{liu-etal-2023-ccl23-eval-ren,
    title = "{CCL}23-Eval 任务6系统报告:基于{CLS}动态加权平均和数据增强的电信网络诈骗案件分类(System Report for {CCL}23-Eval Task 6:::Classification of Telecom {I}nternet Fraud Cases Based on {CLS} Dynamic Weighted Average and Data Augement)",
    author = "Liu, Tianyun  and
      Zhang, Xinghua  and
      Song, Mengxiao  and
      Liu, Tingwen",
    editor = "Sun, Maosong  and
      Qin, Bing  and
      Qiu, Xipeng  and
      Jiang, Jing  and
      Han, Xianpei",
    booktitle = "Proceedings of the 22nd Chinese National Conference on Computational Linguistics (Volume 3: Evaluations)",
    month = aug,
    year = "2023",
    address = "Harbin, China",
    publisher = "Chinese Information Processing Society of China",
    url = "https://preview.aclanthology.org/author-degibert/2023.ccl-3.19/",
    pages = "179--183",
    language = "zho",
    abstract = "``电信网络诈骗领域的案件分类作为文本分类的一项落地应用,其目的是为相关案件进行智能化的分析,有助于公安部门掌握诈骗案件的特点,针对性的预防、制止、侦查。本文以此问题为基础,从模型设计、训练过程、数据增强三个方面进行了研究,通过CLS动态加权平均、Multi-Sample Dropout、对抗训练FGM、回译等方法显著提升了模型对诈骗案件描述的分类性能。''"
}