avatar

Hexuan Yu

于鹤旋
Pronunciation: huh-SHWEN yoo
Ph.D. Candidate in Computer Science
Virginia Tech
📍 Arlington, VA
✉️ hexuanyu@vt.edu
Research Interests
AI, LLM & Agent Security
Applied Cryptography & Privacy
Network, Systems & Communication Security

About Me

I am a Ph.D. Candidate in Computer Science at Virginia Tech, advised by Prof. Wenjing Lou in the Complex Networks and Security Research (CNSR) group. I am a security and privacy researcher working across AI, networked, and computing systems. My work spans LLM and agent security and privacy, including agent privacy, workflow anomaly detection and monitoring, and trustworthy AI training and inference; applied cryptography and privacy, including anonymous authentication, privacy-preserving computation, and hardware-backed roots of trust and attestation (TEEs, TPMs, and SIMs); and network and communication security, including security and privacy in cellular and 5G networks, spam and scam defense, network monitoring, V2X communications, and satellite networks.

A recurring theme in my research is studying security and privacy across interconnected components and layers, and designing practical protocols and mechanisms to address these challenges. I am particularly interested in cross-layer problems, where security and privacy depend on interactions among multiple components, protocols, trust boundaries, and real-world system constraints. My work combines the analysis of these interactions with the design and evaluation of practical solutions that integrate appropriate security mechanisms across different parts of the system.

Before joining Virginia Tech, I received my M.S. in Computer Science from Arkansas State University and my B.S. in Communication Engineering from Beijing University of Posts and Telecommunications (BUPT). During my Ph.D., I was also a Research Intern at Microsoft Research, where I worked on security and privacy for agentic AI systems, mentored by Dr. Betül Durak and Dr. Satya Lokam.

News

  • [Sept. 2026] Our work on radar-enabled emergency communication was accepted to IEEE MILCOM 2026.
  • [Aug. 2026] SKYNET, our work on workflow anomaly detection for agentic AI, was accepted to ACM MobiHoc 2026.
  • [May–Aug. 2026] I spent the summer as a Research Intern with the Cryptography Group at Microsoft Research (Redmond, WA), collaborating with the Windows team on security and privacy for agentic AI systems.
  • [June 2026] SpexPay, our work on privacy-preserving spectrum sharing, was accepted to ESORICS 2026.
  • [May 2026] Our work on auditable and verifiable AI fine-tuning and inference was accepted to USENIX Security 2026.
  • [Feb. 2026] I presented ANONYCALL, our work on privacy-preserving cellular calling, at NDSS 2026 in San Diego, CA; the artifact received all three NDSS Artifact Evaluation badges — Available, Functional, and Reproduced.
  • [Sept. 2025] I presented our work on verifiable monitoring for Open RAN operations virtually at IEEE CNS 2025 in Avignon, France.
  • [June 2024] I was invited to present our work on privacy-preserving spectrum allocation virtually at the 2024 USN–USMC Electromagnetic Spectrum Summit in Norfolk, VA.
  • [May 2024] I presented Pri-Share, our work on privacy-preserving spectrum allocation, at IEEE DySPAN 2024 in Washington, DC.
  • [Feb. 2024] I presented AAKA, our work on privacy-preserving cellular authentication, at NDSS 2024 in San Diego, CA.
  • [Aug. 2023] I presented UCBlocker, our work on combating spam and scam calls, at USENIX Security 2023 in Anaheim, CA.

Selected Publications (for the complete list, see my Google Scholar profile)

  1. MINIM: Privacy-Aware Minimal View for Agents via Trusted Local Sanitization [Paper]
    Hexuan Yu, Chaoyu Zhang, Heng Jin, Shanghao Shi, Ning Zhang, Y. Thomas Hou, Wenjing Lou
    International Conference on Machine Learning (ICML), 2026.
  2. ANONYCALL: Enabling Native Private Calling in Mobile Networks [Paper]
    Hexuan Yu, Chaoyu Zhang, Yang Xiao, Angelos D. Keromytis, Y. Thomas Hou, Wenjing Lou
    Network and Distributed System Security Symposium (NDSS), 2026. 265/1,481 papers accepted (17.9%).
    NDSS Artifact Evaluation badge: Available, Functional, Reproduced
  3. V-PASS: Sybil-Resistant Pseudonym Self-Provisioning for V2X [Paper]
    Hexuan Yu, Md Mohaimin Al Barat, Shaoyu Li, Md Hasan Shahriar, Yang Xiao, Panagiotis Papadimitratos, Y. Thomas Hou, Wenjing Lou
    ACM Conference on Security and Privacy in Wireless and Mobile Networks (WiSec), 2026. 31/145 papers accepted (21.4%).
  4. Skynet: Workflow-Level Anomaly Detection for Agentic AI via Semantic and Structural Modeling [Paper]
    Chaoyu Zhang*, Hexuan Yu*, Heng Jin, Shanghao Shi, Ning Zhang, Yi Shi, Yulia R. Gel, Y. Thomas Hou, Wenjing Lou * Equal contribution
    ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc), 2026 (To appear). 60/260 papers accepted (23.1%).
  5. Trusting What You Cannot See: Auditable Fine-Tuning and Inference for Proprietary AI [Paper]
    Heng Jin, Chaoyu Zhang, Hexuan Yu, Shanghao Shi, Ning Zhang, Y. Thomas Hou, Wenjing Lou
    USENIX Security Symposium, 2026. 362/2,750 papers accepted (13.2%).
  6. FC-GUARD: Enabling Anonymous yet Compliant Fiat-to-Cryptocurrency Exchanges [Paper]
    Shaoyu Li, Hexuan Yu, Md Mohaimin Al Barat, Yang Xiao, Y. Thomas Hou, Wenjing Lou
    IEEE International Conference on Computer Communications (INFOCOM), 2026. 327/1,740 papers accepted (18.8%).
  7. SpexPay: A Privacy-Preserving Pay-As-You-Go System for Dynamic Spectrum Sharing [Paper]
    Md Mohaimin Al Barat, Hexuan Yu, Shaoyu Li, Yang Xiao, Yi Shi, Eric W. Burger, Y. Thomas Hou, Wenjing Lou
    European Symposium on Research in Computer Security (ESORICS), 2026.
  8. Enabling Emergency Communication via Semantic Radar-Centric ISAC [Paper]
    Md Mohaimin Al Barat, Chaoyu Zhang, Hexuan Yu, Kai Zeng, Y. Thomas Hou, Wenjing Lou
    IEEE Military Communications Conference (MILCOM), 2026 (To appear).
  9. Privacy-Preserving Split Learning for Federated LLM Fine-Tuning [Paper]
    Heng Jin, Chaoyu Zhang, Hexuan Yu, Wenjing Lou, Y. Thomas Hou
    arXiv preprint, 2026.
  10. ClawCoin: An Agentic AI-Native Cryptocurrency for Decentralized Agent Economies [Paper]
    Shaoyu Li, Chaoyu Zhang, Hexuan Yu, Y. Thomas Hou, Wenjing Lou
    2026 Nanyang Blockchain Conference (NBC), 2026.
  11. Closing the Visibility Gap: A Monitoring Framework for Verifiable Open RAN Operations [Paper]
    Hexuan Yu, Md Mohaimin Al Barat, Yang Xiao, Y. Thomas Hou, Wenjing Lou
    IEEE Conference on Communications and Network Security (CNS), 2025.
  12. StarCast: A Secure and Spectrum-Efficient Group Communication Scheme for LEO Satellite Networks [Paper]
    Chaoyu Zhang, Hexuan Yu, Shanghao Shi, Shaoyu Li, Yi Shi, Eric Burger, Y. Thomas Hou, Wenjing Lou
    IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN), 2025.
  13. Enabling Trustworthy Federated Learning via Remote Attestation for Mitigating Byzantine Threats [Paper]
    Chaoyu Zhang, Heng Jin, Shanghao Shi, Hexuan Yu, Sydney Johns, Y. Thomas Hou, Wenjing Lou
    IEEE Military Communications Conference (MILCOM), 2025.
  14. AAKA: An Anti-Tracking Cellular Authentication Scheme Leveraging Anonymous Credentials [Paper]
    Hexuan Yu, Changlai Du, Yang Xiao, Angelos D. Keromytis, Chonggang Wang, Robert Gazda, Y. Thomas Hou, Wenjing Lou
    Network and Distributed System Security Symposium (NDSS), 2024. 140/682 papers accepted (20.5%).
  15. Pri-Share: Enabling Inter-SAS Privacy Protection via Secure Multi-Party Spectrum Allocation [Paper]
    Hexuan Yu, Shanghao Shi, Yi Shi, Eric W. Burger, Y. Thomas Hou, Wenjing Lou
    IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN), 2024.
  16. UCBlocker: Unwanted Call Blocking Using Anonymous Authentication [Paper]
    Changlai Du, Hexuan Yu, Yang Xiao, Y. Thomas Hou, Angelos D. Keromytis, Wenjing Lou
    USENIX Security Symposium, 2023.
  17. Mobile Tracking in 5G and Beyond Networks: Problems, Challenges, and New Directions [Paper]
    Changlai Du, Hexuan Yu, Yang Xiao, Wenjing Lou, Chonggang Wang, Robert Gazda, Y. Thomas Hou
    IEEE International Conference on Mobile Ad Hoc and Smart Systems (MASS), 2022.

Professional Service

Conference Service

Web Chair, IEEE Conference on Communications and Network Security (CNS), 2026

Conference Reviewing

  • IEEE International Conference on Computer Communications (INFOCOM), 2025
  • IEEE Military Communications Conference (MILCOM), 2024
  • European Symposium on Research in Computer Security (ESORICS), 2022–2023
  • IEEE Conference on Communications and Network Security (CNS), 2022

Journal Reviewing

  • IEEE Transactions on Information Forensics and Security (TIFS)
  • IEEE/ACM Transactions on Networking (ToN)
  • IEEE Transactions on Dependable and Secure Computing (TDSC)

External Reviewing

  • IEEE Symposium on Security and Privacy (S&P), 2024–2026
  • ACM Asia Conference on Computer and Communications Security (AsiaCCS), 2026
  • ACM Conference on Security and Privacy in Wireless and Mobile Networks (WiSec), 2023–2026
  • ACM Conference on Computer and Communications Security (CCS), 2023
  • USENIX Security Symposium, 2023
  • IEEE International Conference on Computer Communications and Networks (ICCCN), 2023