网络与信息安全学报2025，Vol.11Issue(2)：161-174,14.DOI:10.11959/j.issn.2096-109x.2025026
基于标点符号插入的中文对抗文本生成方法
Chinese adversarial text generation method based on punctuation insertion
摘要
Abstract
The susceptibility of natural language processing models to adversarial texts has been a significant con-cern.Current methods for generating adversarial texts in Chinese were mainly based on replacing characters with visually similar or homophonic ones.However,when faced with robust pre-trained models,these methods led to in-creased perturbations in adversarial texts,resulting in reduced fluency and readability,and thus generating low-quality adversarial texts.Moreover,symbol insertion methods used in English adversarial texts were not entirely ap-plicable to Chinese.Additionally,in a black-box scenario,the lack of prior knowledge made it difficult to generate high-quality adversarial texts.A punctuation-based method for generating adversarial texts for Chinese text classifi-cation tasks was proposed.Under a black-box setting,a novel part-of-speech importance calculation was utilized and combined with punctuation insertion to design a character-level perturbation approach suitable for Chinese,achieving the generation of adversarial texts.Experiments were conducted,and the results demonstrated that for text classification tasks,the proposed method significantly improved the attack success rate on LSTM and BERT models trained with two real-world datasets.Furthermore,the method successfully avoided direct destruction of the original sentences and maintained the original meaning.In the tests,a semantic similarity of up to 97%was achieved,which was significantly better than the baseline methods.关键词
中文文本分类/对抗文本生成/黑盒攻击Key words
Chinese text classification/adversarial text generation/black-box attack分类
计算机与自动化引用本文复制引用
张谦,闫巧..基于标点符号插入的中文对抗文本生成方法[J].网络与信息安全学报,2025,11(2):161-174,14.基金项目
国家自然科学基金(61976142) （61976142）
深圳科技计划项目(JCYJ20210324093609025) The National Natural Science Foundation of China(61976142),Shenzhen Science and Technology Plan-ning Project(JCYJ20210324093609025) （JCYJ20210324093609025）
