@inproceedings{liang-yu-lan-etal-2025-ji,
    title = "基于多模型协同的儿童互联网新闻风险管理与价值观引导框架",
    author = "梁宇蓝, 梁宇蓝  and
      王悦, 王悦  and
      Yu, Dong  and
      Liu, Pengyuan  and
      Kang, Chen",
    editor = "Sun, Maosong  and
      Duan, Peiyong  and
      Liu, Zhiyuan  and
      Xu, Ruifeng  and
      Sun, Weiwei",
    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-1.30/",
    pages = "402--421",
    abstract = "随着互联网在儿童群体中的广泛普及,新闻内容的{''}毒性遗留{''}与价值观缺失已成为亟待解决的安全挑战。本文提出了一种多模型协同的儿童新闻改写框架(CRV-LLM),旨在从词汇、事件、标题和价值观四个维度,对原始新闻文本进行深度风险识别与精准改写。CRV-LLM集成了四个轻量化风险检测模型和R1-Distill-Qwen-32B改写模型,通过模型间的协同与反馈,能够在保证儿童可读性的前提下,有效剔除潜在有害信息并植入积极价值引导。实验结果表明,CRV-LLM框架在安全性、教育性等核心指标上优于主流模型,且推理效率提升62{\%},为儿童互联网内容安全管理提供了一种高效、可扩展的技术方案。"
}