Chunyi Zhou

Safeguarding Visual Privacy in Dataset Distillation: Robust Initialization via Augmentation

When Agents “Misremember” Collectively: Exploring the Mandela Effect in LLM-based Multi-Agent Systems

FedGraph: Defending Federated Large Language Model Fine-Tuning Against Backdoor Attacks via Graph-Based Aggregation

Arcueid: Multi-trigger Cloud Shaping for Unified Backdoor Attack Paradigms

MASpi: A Unified Environment for Evaluating Prompt Injection Robustness in LLM-Based Multi-Agent Systems

Angel or Demon: Investigating the Plasticity-Enhanced Strategies' Impact on Backdoor Threats in Deep Reinforcement Learning

Draining Your Account: A Stealthy Attack on API Billing in Multi-Agent Systems