Kun Zhai — accepted AI-conference papers

2026

SkillTrojan: Backdoor Attacks on Skill-Based Agent Systems

Yunhao Feng, Yifan Ding, Yingshui Tan, Boren Zheng, Yanming Guo, Xiaolong Li +3

ICML 2026poster

Skill-based agent systems tackle complex tasks by composing reusable skills, improving modularity and scalability while introducing a largely unexamined security attack surface. We propose \textbf{SkillTrojan}, a backdoor attack that targets skill implementations rather than model parameters or trai…

Cited by 0 Source ↗Scholar ↗

2025

FedCAda: Adaptive Client-Side Optimization for Accelerated and Stable Federated Learning

Liuzhi Zhou, Yu He, Kun Zhai, Xiang Liu, Sen Liu, Xingjun Ma +2

ICASSP 2025accepted

Federated learning (FL) enables collaborative model training across distributed clients while preserving data privacy. However, achieving both acceleration and stability, particularly on the client side, remains a challenge. In this paper, we introduce FedCAda, an adaptive algorithm that leverages a…

Cited by 0 Source ↗Scholar ↗