计算机与数字工程2025，Vol.53Issue(11)：3144-3148,3227,6.DOI:10.3969/j.issn.1672-9722.2025.11.027
面向中文的字符级对抗样本生成方法
Chinese Character-level Adversarial Example Generation Approach
摘要
Abstract
Adversarial examples are an important means to measure the robustness of deep learning models.In the field of Chi-nese text adversarial example generation,text adversarial examples have the problems of low coverage situations,weakness attack effect and low quality.To address the above problems,a character-level text adversarial example generation approach SCGA(Simi-lar Character Generation Adversarial Example)is proposed,SCAG constructs a phonetic-form similarity dictionary by calculating the similarity of pronunciation and shape.This method effectively locates important word positions,using similar characters to re-place the original examples,and finally performing an adversarial example attack under black-box.This paper verifies the method's effectiveness on several datasets compared with several methods.This method has high quality and can effectively mislead the classi-fication results of the model.关键词
深度学习/文本对抗样本生成/中文文本分类模型/黑盒攻击/相近字典Key words
deep learning/textual adversarial example generation/Chinese text classification model/black-box attack/nearby dictionaries分类
信息技术与安全科学引用本文复制引用
胡汉平,张洁..面向中文的字符级对抗样本生成方法[J].计算机与数字工程,2025,53(11):3144-3148,3227,6.基金项目
国家重点研发计划(编号:2018YFB1500902)资助. （编号:2018YFB1500902）
