@inproceedings{konglulu-etal-2025-ccl25,
    title = "{CCL}25-Eval任务11系统报告:基于大模型微调的汉字硬笔书写质量自动评价",
    author = "KongLulu, KongLulu  and
      Zan, Hongying  and
      Song, Jinwang  and
      LiuHaixin, LiuHaixin  and
      Li, Yifan  and
      Luo, Zhewei",
    editor = "Lin, Hongfei  and
      Li, Bin  and
      Tan, Hongye",
    booktitle = "Proceedings of the 24th {C}hina National Conference on Computational Linguistics ({CCL} 2025)",
    month = aug,
    year = "2025",
    address = "Jinan, China",
    publisher = "Chinese Information Processing Society of China",
    url = "https://preview.aclanthology.org/ingest-ccl/2025.ccl-2.51/",
    pages = "430--436",
    abstract = "本技术报告探讨了通过微调本地视觉语言模型,实现汉字硬笔书写质量自动评价的技术方案。针对传统评价方法难以提供准确性反馈的问题,我们团队采用精心设计的prompt并结合微调的方式构建了一个高效的汉字硬笔书写质量自动评价系统。我们采用Qwen2.5-VL-7B-Instruct模型作为基础,通过LoRA微调技术实现了汉字书写质量等级分类(子任务一)和个性化评语生成(子任务二)的功能。系统地融合了视觉特征分析与语言生成能力,在训练过程中采用了梯度检查点、BF16混合精度训练等技术优化显存使用,并设计了针对性的损失函数和评估指标。实验结果表明,我们的方法能够有效实现汉字书写质量的细粒度评价。"
}