Mengdi Yang
Papers
1
Total Citations
5
H-Index
1
About
Mengdi Yang is a security and privacy researcher whose work critically examines the autonomy and tracking capabilities embedded in modern devices. Her research focuses on the intersection of computer vision, embedded systems, and adversarial machine learning, with a particular emphasis on exposing hidden surveillance and tracking functionalities in consumer and industrial hardware. Yang’s most cited work, “A Black Box Approach to Inferring, Characterizing, and Breaking Native Device Tracking Autonomy” (2020), introduces a novel methodology for reverse-engineering autonomous tracking systems without access to proprietary source code. By treating these systems as black boxes, she demonstrates how to infer their tracking logic, characterize their behavioral boundaries, and ultimately develop techniques to disrupt or evade them. This contribution has significant implications for privacy protection, especially as autonomous devices—from drones to smart cars—become ubiquitous. With over 5 citations, this paper has influenced subsequent research in adversarial attacks on vision-based tracking and has been recognized for its practical, system-level approach. Yang’s work empowers users and developers to understand and challenge the autonomy of devices that increasingly shape our physical and digital environments.
Research Focus
Key Achievements
Top Papers
- 1