［1］Wang X, Cao S, Zheng K, et al. Supervised character resemble substitution personality adversarial method［J］. Electronics, 2023, 12(4): 869［2］靳文京, 卜哲, 秦博阳. 基于序列生成对抗网络的智能模糊测试方法［J］. 信息安全研究, 2024, 10(6): 490497［3］Xue J, Zheng M, Hua T, et al. Trojllm: A blackbox trojan prompt attack on large language models［J］. Advances in Neural Information Processing Systems, 2023, 36(11): 6566565677［4］Carlini N, Tramer F, Wallace E, et al. Extracting training data from large language models［EBOL］. 2020 ［20250717］. https:xueshu.baidu.comusercenterpapershow?paperid=151h0jc0xn1e02u0pp080tc05m288830&site=xueshu_se &hitarticle=1［5］ZanellaBeguelin S, Tople S, Paverd A, et al. Greybox extraction of natural language models［EBOL］. 2021［20250717］. https:xueshu.baidu.comusercenterpapershow?paperid=1t4c0p20ee4x08y0nc0y0e60qj557159&site=xueshu_se&hitarticle=1［6］王子晨, 汤艳君, 潘奕扬. 面向取证的网络攻击者溯源分析技术研究综述［J］. 信息安全研究, 2024, 10(4): 302310［7］Burford K G, Itzkowitz N G, Ortega A G, et al. Use of generative AI to identify helmet status among patients with micromobilityrelated injuries from unstructured clinical notes［J］. JAMA Network Open, 2024, 7(8): e2425981e2425981［8］顾芳铭, 况博裕, 许亚倩, 等. 面向自动驾驶感知系统的对抗样本攻击研究综述［J］. 信息安全研究, 2024, 10(9): 786794［9］Carlini N, Tramer F, Wallace E, et al. Extracting training data from large language models［EBOL］. 2020 ［20250717］. https:xueshu.baidu.comusercenterpapershow?paperid=151h0jc0xn1e02u0pp080tc05m288830&site=xueshu_se &hitarticle=1［10］Ilyas A, Engstrom L, Athalye A, et al. Blackbox adversarial attacks with limited queries and information［EBOL］. 2018 ［20250717］. https:xueshu.baidu.comusercenterpapershow?paperid=45d483db0b356e0b7b16bb302f02f38 2&site=xueshu_se |
