Zan Zhou

About

Zan Zhou is a leading researcher at the forefront of security and privacy in distributed machine learning, with a particular focus on the Internet of Robotic Things (IoRT) and large vision-language models (LVLMs). Their work addresses critical vulnerabilities in federated learning systems, where heterogeneous cross-silo environments are increasingly susceptible to attacks. Zhou’s major contribution, "Safeguarding Privacy and Integrity of Federated Learning in Heterogeneous Cross-Silo IoRT Environments: A Moving Target Defense Approach" (2024, 14 citations), pioneers a dynamic defense mechanism that thwarts adversarial manipulation by continuously shifting the attack surface, thereby preserving both data privacy and model integrity in complex robotic networks. Building on this, Zhou’s "SecFFT: Safeguarding Federated Fine-Tuning for Large Vision Language Models Against Covert Backdoor Attacks in IoRT Networks" (2024, 7 citations) introduces a novel framework to detect and neutralize stealthy backdoor threats during fine-tuning, a critical step for deploying LVLMs in smart city and industrial applications. With a growing citation impact and a focus on real-world IoRT deployments—spanning factories, power grids, and transportation—Zhou’s work is essential for ensuring the safe, trustworthy integration of AI into autonomous systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Safeguarding Privacy and Integrity of Federated Learning in Heterogeneous Cross-Silo IoRT Environments: A Moving Target Defense Approach

   14 citations · 2024
2. 2
   SecFFT: Safeguarding Federated Fine-Tuning for Large Vision Language Models Against Covert Backdoor Attacks in IoRT Networks

   7 citations · 2024

Key Collaborators

Contact & Links