Jul 2026
FedLSG: LLM-Enhanced Semantic Calibration for Federated Graph Backdoor Defense
FedLSG, the first framework that integrates large language models (LLMs) into federated graph backdoor defense, introduces a graph and behavior to text grounding scheme that transforms local graph structures and client update behaviors into semantically rich natural language representations.
Chenyu Zhou, Ya-Bin Peng, Wei Huang et al.
· arXiv.org · 0 citations