北京邮电大学网络空间安全博士，新西兰奥克兰大学CSC国家公派博士联培一年。

邮箱：zero10walker@ysu.edu.cn; 办公地点：信息馆1号楼。

博士期间主持课题：

1. 中央网信办与蚂蚁科技集团，网络安全学院学生创新资助计划，拆分学习攻防，2022-10 至 2023-10

博士期间参与课题：

1. 海量终端低时延高可靠多域资源协同与调配技术: 2021.12-2024.11，重点研发计划课题，2021YFB2900302

2. 面向 6G 高效数据安全流通的可重构分布式学习机理: 2024.01-2027.12，国家自然科学基金面上项目，62371059

3. 数据共享服务网络架构优化与安全方案研究: 2023.01-2023.12，北邮-中移动联合创新科研项目，CMYJY-202300014

近五年论文发表：

1. Junlin Liu, Xinchen Lyu, Longfei Zheng, Chenshan Ren, and Qimei Cui. “PPoison: A Pluggable Poisoning Attack Against Distributed Training of Split Learning,” Future Generation Computer Systems. Elsevier, 2026.（中科院二区）

2. Junlin Liu, Xinchen Lyu, Chenshan Ren, and Qimei Cui. “Targeted Poisoning Attacks Against Vertical Federated Learning Via Embedding Manipulation,” IEEE Transactions on Dependable and Secure Computing. IEEE, 2025.（中科院二区top，CCF-A）

3. Junlin Liu, Xinchen Lyu, Chenshan Ren, and Qimei Gui. “Crafting More Transferable Adversarial Examples via Quality-aware Transformation Combination.” IEEE Transactions on Multimedia. IEEE, 2025.（中科院一区top，CCF-A）

4. Junlin Liu, Xinchen Lyu, Qimei Cui and Xiaofeng Tao. “Similarity-based Label Inference Attack against Training and Inference of Split Learning.” IEEE Transactions on Information Forensics and Security. IEEE, 2024.（中科院一区top，CCF-A）

5. Junlin Liu, and Xinchen Lyu. “Boosting the Transferability of Adversarial Examples via Local Mixup and Adaptive Step Size.” ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2025.（CCF-B会议）

6. Junlin Liu, and Xinchen Lyu. “Distance-Based Online Label Inference Attacks Against Split Learning.” ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2023.（CCF-B会议）

7. Xinchen Lyu, Junlin Liu, Chenshan Ren, and Guoshun Nan. “Security-communication-computation tradeoff of split decisions for edge intelligence.” IEEE Wireless Communications. IEEE, 2023.（中科院二区）

邮箱 : 83042b18d96accef5a7baf2a34b0510bc9933afce9f483b5062242014a1f3e973e30b6fafe7850cd23d16d84a3f6d49a5dbcaeee0076785990f8c2950edefb9e237c9bf45862788c5c04882d0339e386dd6b43552f4c83037a38ec70b2cd307757cfdce564a98a6f3a623a911f3c7fb1d86fb25e96603667174d82197c7bc453