[CySeP’26] Special thanks to all volunteers (B. Bilas, G. Bujor, Y. Cheng, K. Domkina, Z. Jia, B. Lagerros, A. Wigren, M. Zhang) and all participants! Hope to see many of you next year! #NSS_KTH#Security#Privacy
Photos by Zhichao Zhou
@NetSysSecKTH@KTHuniversity
📢Attending ACM SAC'26? Check out @ShengLiuKTH’s presentation on “DEFEND: Poisoned Model Detection and Malicious Client Exclusion Mechanism for Secure Federated Learning-based Road Condition Classification”, joint work w. @PanosPapadimitr (S2RAI: Safe, Secure and Robust AI Track)
This paper shows that FL-based intelligent transportation systems are vulnerable to targeted label-flipping attacks. We introduce DEFEND, a countermeasure that utilizes neuron-wise analysis to detect poisoned model and leverages adaptive rating to exclude malicious clients.
This paper shows that FL-based intelligent transportation systems are vulnerable to targeted label-flipping attacks. We introduce DEFEND, a countermeasure that utilizes neuron-wise analysis to detect poisoned model and leverages adaptive rating to exclude malicious clients.