唐海龙，（安徽安庆人），2023年-2026年就读于合肥工业大学计算机与信息学院，攻读硕士学位。研究方向为联邦学习。硕士大论文题目：集群联邦学习中面向后门攻击的防御策略研究。现任职于小米。

学业简历：安庆师范大学（本科）合肥工业大学（研究生）

学术成果：

(1)Tang Hailong, Shi Lei, Zhu Yingfei, Gu Cheng, Xu Juan. The Defense Against Backdoor Attacks Using Trigger Inversion and Data Augmentation in Clustered Federated Learning[A]. Wireless Artificial Intelligent Computing Systems and Applications[C]. Singapore: Springer Nature Singapore, 2025: 331-343.(CCF C类会议收录，本人一作)

(2)Gu Cheng, Shi Lei, Liu Binbin, Tang Hailong, Xu Juan. Adaptive Privacy Defense Against Category Inference Attack in Clustered Federated Learning: Balancing Security and Model Performance[A]. Wireless Artificial Intelligent Computing Systems and Applications[C]. Singapore: Springer Nature Singapore, 2025: 245-255.(CCF C类会议收录)

(3)Shi Lei, Gu Cheng, Fan Yuqi, Tang Hailong and Zhu Yingfei. Fisher-Driven Privacy Preservation Against Category Inference Attacks in Federated Learning[J]. High-Confidence Computing, 2026: 100399.(CCF C类期刊收录)

(4)石雷，许浩，唐海龙，张洋，钱定军，顾程，“针对集群联邦学习聚类过程的安全验证方法、终端及介质”，中国发明专利，授权号：CN119646811B，2025

(5)石雷，朱迎飞，钱定军，顾程，唐海龙，张洋，李泽鹏，“一种攻击模型数据集的获取方法及攻击模型数据集的应用”，中国发明专利，专利号：ZL202610357011.8，2026.03.23（待授权）

When one chapter ends, another begins.

2026/6/3