FedMentor: 域感知差分隐私用于心理健康领域异构联邦大语言模型
FedMentor: Domain-Aware Differential Privacy for Heterogeneous Federated LLMs in Mental Health
AI总结：
FedMentor通过整合LoRA和域感知差分隐私，实现对心理健康领域异构联邦大语言模型的隐私保护微调，提升安全性并保持模型效用。
摘要：
在敏感领域（如心理健康）中，保护大型语言模型（LLMs）的隐私需要在严格保密性与模型效用和安全之间取得平衡。我们提出了FedMentor，一个联邦微调框架，整合了低秩适应（LoRA）和域感知差分隐私（DP），以满足每个领域的隐私预算，同时保持性能。每个客户端（领域）应用一个自定义的DP噪声比例，与其数据敏感性成正比，而服务器则在效用低于阈值时自适应减少噪声。在三个心理健康数据集上的实验中，我们证明FedMentor在无隐私标准联邦学习（FL）的基础上提高了安全性，安全输出率提高了多达3个百分点，降低了毒性，同时保持效用（BERTScore F1和ROUGE-L）在非隐私基准的0.5%以内，并接近集中化上限。该框架可扩展到具有高达17亿参数的主干模型上，在单块GPU客户端上运行，每轮通信需求小于173MB。FedMentor展示了在医疗和其他敏感领域中安全微调LLMs的实用方法。
英文摘要：
Privacy-preserving adaptation of Large Language Models (LLMs) in sensitive domains (e.g., mental health) requires balancing strict confidentiality with model utility and safety. We propose FedMentor, a federated fine-tuning framework that integrates Low-Rank Adaptation (LoRA) and domain-aware Differential Privacy (DP) to meet per-domain privacy budgets while maintaining performance. Each client (domain) applies a custom DP noise scale proportional to its data sensitivity, and the server adaptively reduces noise when utility falls below a threshold. In experiments on three mental health datasets, we show that FedMentor improves safety over standard Federated Learning (FL) without privacy, raising safe output rates by up to three points and lowering toxicity, while maintaining utility (BERTScore F1 and ROUGE-L) within 0.5% of the non-private baseline and close to the centralized upper bound. The framework scales to backbones with up to 1.7B parameters on single-GPU clients, requiring < 173 MB of communication per-round. FedMentor demonstrates a practical approach to privately fine-tune LLMs for safer deployments in healthcare and other sensitive fields.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
