发表机构
Southeast University; Purple Mountain Laboratories; Institute of AI for Industries; Chinese Academy of Sciences（东南大学; 紫金山实验室; 人工智能产业研究院; 中国科学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对联邦图神经网络易受后门攻击问题，提出FedLSG框架，将大语言模型集成防御，通过图与行为到文本的转换及轻量级师生架构，在不损图完整性时显著提升对后门攻击的抵抗力。
AI 中文摘要
联邦图神经网络极易受到后门攻击，而现有防御通常依赖缺乏语义理解的基于规则的方法，易受隐秘触发因素影响且对良性结构有害。为此，我们提出了FedLSG，这是首个将大语言模型集成到联邦图后门防御中的框架。它引入了一种将图和行为转换为文本基础的方案，采用轻量级师生架构。实验表明，FedLSG在不损害图完整性的情况下显著提高了对后门攻击的抵抗力。
英文摘要
Federated Graph Neural Networks (FedGNNs) are highly vulnerable to backdoor poisoning, yet existing defenses typically rely on rule-based approaches that lack semantic understanding, making them vulnerable to stealthy triggers and harmful to benign structures. To solve this, we present FedLSG, the first framework that integrates large language models (LLMs) into federated graph backdoor defense. FedLSG introduces a graph and behavior to text grounding scheme that transforms local graph structures and client update behaviors into semantically rich natural language representations. The framework further adopts a lightweight student-teacher architecture. On the server side, a full scale LLM serves as a teacher, providing global contextual guidance and evaluating client updates during aggregation to identify potentially malicious participants. On the client side, a LoRA-based student is maintained to perform semantic reasoning, to suppress the influence of edges associated with backdoor triggers. By enabling semantic interpretation of both graph patterns and client behaviors, the framework adaptively incorporates rule-based signals into message passing and client aggregation for defense. Experiments demonstrate that FedLSG significantly improves resistance to backdoor attacks without compromising graph integrity.
