Meng, Weizhi

How Breakable Is Privacy: Probing and Resisting Model Inversion Attacks in Collaborative Inference

2025 · arXiv (Cornell University)

Collaborative inference (CI) improves computational efficiency for edge devices by transmitting intermediate features to cloud models. However, this process inevitably exposes feature representations to model inversion attacks (MIAs), enabling unauthorized data reconstruction. Despite extensive research, …