2019年9月起在西安电子科技大学进行硕博连读,师从沈玉龙教授,并于2026年获计算机科学与技术专业博士学位。期间,他于2024年至2025年赴奥克兰大学访学,合作导师为Giovanni Russello教授。他的本科毕业于河南工业大学软件工程专业。他将于2026年8月加入澳门城市大学担任教职。他的研究方向涵盖安全多方计算(MPC)、联邦学习、隐私保护机器学习、大小模型协同安全以及应用密码学。

Anxiao Song is currently pursuing a Ph.D. in Computer Science and Technology at Xidian University under Prof. Yulong Shen, and was a visiting researcher at the University of Auckland (2024–2025) under Prof. Giovanni Russello, funded by CSC. He received his B.E. from Henan University of Technology. He will join City University of Macau as a faculty member in August 2026. His research interests include secure MPC, federated learning, privacy-preserving ML, collaborative security for small and large models, and applied cryptography.

🎓 教育经历 / Education

📝 代表性论文 / Selected Publications

  • Anxiao Song, Shujie Cui, Ke Cheng, Jianli Bai, Qifan Wang, Shangqi Lai, Giovanni Russello, Yulong Shen. FSS-DT: Secure Division-free Decision Tree Training and Inference via Function Secret Sharing[J]. IEEE Transactions on Dependable and Secure Computing, 2026. (已接收 / Accepted, CCF-A)

  • Anxiao Song, Ke Cheng, Jiaxuan Fu, et al. Private Learning for Vertical Decision Trees: A Secure, Accurate, and Fast Realization[J]. IEEE Transactions on Dependable and Secure Computing, vol. 22, no. 6, pp. 6088-6103, Nov.-Dec. 2025. (CCF-A)
    [网页]

  • Anxiao Song, Shujie Cui, Jianli Bai, et al. Guard-GBDT: Efficient Privacy-Preserving Approximated GBDT Training on Vertical Dataset[C]. 2025 28th International Symposium on Research in Attacks, Intrusions and Defenses (RAID). IEEE, 2025: 741-756. (CCF-B)
    [网页]

  • Anxiao Song, Haoshu Li, Ke Cheng, et al. Guard-FL: An UMAP-assisted robust aggregation for federated learning[J]. IEEE Internet of Things Journal, 2024, 11(16): 27557-27570. (中科院一区)
    [网页]

  • Anxiao Song, Tao Zhang, Ke Cheng, et al. Byzantine-Robust Federated Learning Framework via a Server-Client Defense Mechanisms[J]. IEEE Internet of Things Journal, vol. 12, no. 14, pp. 29073-29088, 15 July 2025. (中科院一区)
    [网页]

  • Ke Cheng, Yuheng Xia, Anxiao Song, et al. Mosformer: Maliciously Secure Three-Party Inference Framework for Large Transformers[C]. Proceedings of the 2025 ACM SIGSAC Conference on Computer and Communications Security (CCS ‘25). ACM, 2025: 4124–4138. (CCF-A)
    [网页]

🏅 荣誉奖项 / Honors

  • GitCode Top 10 开源项目奖 (NssMPClib)

🔬 审稿服务 / Review Service

  • IEEE Transactions on Sustainable Computing (TSC)
  • IEEE Transactions on Dependable and Secure Computing (TDSC)
  • IEEE Transactions on Information Forensics and Security (TIFS)

💼 工作经历 / Experience

  • 2026.08 - 至今, 澳门城市大学 (City University of Macau), 助理教授