ZHENGYUAN JIANG

Zhengyuan Jiang is a Ph.D. candidate in Computer Science and Engineering at the University of South Florida’s Bellini College of Artificial Intelligence, Cybersecurity and Computing. A member of the Security and Privacy Research of AI Systems (SPRAI) Lab, he is advised by Dr. Ning Wang and expects to graduate in Spring 2027.

His research focuses on the security and trustworthiness of distributed learning systems. He studies the security limits of federated and split learning under adaptive adversaries. His work spans practical attacks that uncover vulnerabilities, diagnostic methods that characterize adversarial behavior, and defenses that strengthen system resilience. His work has appeared in ACM Computing Surveys and at ACL, ECAI, and IEEE CNS, with additional work currently under review.

RESEARCH

  • Secure Federated & Distributed LearningBackdoor attacks, poisoning threats, and data-distribution-aware defenses.
  • Trustworthy AI & Information IntegrityReliable machine learning, LLM-driven information integrity, and benchmark development.
  • Adaptive Attacks & Representation ControlProblem-driven research on representation control and adaptive learning dynamics.

EDUCATION

  • Ph.D. in Computer Science & EngineeringUniversity of South Florida · 2021–Present
    Expected Graduation 2027
  • M.S. in Electrical & Computer EngineeringUniversity of Florida · 2019–2021
  • B.S. in Measuring & Control TechnologyShandong University of Science and Technology · 2015–2019

NEWS

  • Reviewed for IEEE Transactions on Information Forensics and Security (TIFS).
  • Launched zhengyuanjiang.com, my personal website.
  • Our paper, “Large Language Models and Social Media Information Integrity: Opportunities, Challenges, and Research Directions,” was published in ACM Computing Surveys.
  • Our paper, “SNAIL: Scheduler-Driven Backdoor Injection in Split Federated Learning via Gradual Perturbation,” was accepted to IEEE CNS 2026.
  • Admitted to candidacy following unanimous approval of the Major Research-Area presentation.
  • Our paper, “Learning from Textual Radiology Reports: A Benchmark Dataset for Coronary CT Angiography,” was accepted to the ACL 2026 Industry Track.
  • Our paper, “BoBa: Boosting Backdoor Detection through Data Distribution Inference in Federated Learning,” was accepted to ECAI 2025.
  • Completed the Ph.D. coursework with a 4.0 GPA and passed the Qualifying Examination.
  • Joined Prof. Ning Wang’s SPRAI Lab to study secure and trustworthy AI systems.