发表机构
State Key Laboratory for Novel Software Technology; School of Artificial Intelligence; School of Computer Science; School of Electronic Science and Engineering; Nanjing University（计算机软件新技术全国重点实验室; 人工智能学院; 计算机科学与技术系; 电子科学与工程学院; 南京大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对联邦微调中统计异质性导致全局适配器难以个性化的问题，提出角色感知框架FedLAFP，结合全局LoRA与私有RandLoRA分支，通过混合系数融合，在四个视觉基准上平均个性化准确率达86.93%，超过最佳基线1.30个百分点。
AI 中文摘要
联邦参数高效微调使客户端能够在无需共享原始数据或通信完整模型的情况下适应预训练模型，但统计异质性使得单一的全局适配器不足以实现个性化预测。现有的个性化方法通常对共享和私有适配采用相同的低秩结构，忽视了它们在聚合和个性化方面的不同需求。我们提出了FedLAFP，一种角色感知框架，它将紧凑的、全局聚合的LoRA分支与客户端私有的、具备全秩能力的RandLoRA分支相结合。共享分支为传递通用知识提供了高效接口，而私有分支将固定的随机低秩基与学习到的缩放系数相结合，以提供表达丰富的客户端特定适应，且无需额外通信。客户端和层特定的混合系数共同融合两个分支，仅共享LoRA参数被交换。一项受控的线性研究支持了这一角色分配：LoRA产生更对齐的客户端更新和更低的聚合误差，而RandLoRA更准确地恢复客户端特定残差。在四个视觉识别基准上的实验表明，FedLAFP始终优于仅本地和联邦LoRA基线，实现了平均个性化准确率$86.93\%$，并超过最佳基线平均值$1.30$个百分点。
英文摘要
Federated parameter-efficient fine-tuning enables clients to adapt pre-trained models without sharing raw data or communicating the full model, but statistical heterogeneity makes a single global adapter insufficient for personalized prediction. Existing personalized methods typically use the same low-rank structure for both shared and private adaptation, overlooking their distinct requirements for aggregation and personalization. We propose FedLAFP, a role-aware framework that couples a compact, globally aggregated LoRA branch with a client-private, full-rank-capable RandLoRA branch. The shared branch provides an efficient interface for transferring common knowledge, whereas the private branch combines fixed random low-rank bases with learned scaling coefficients to provide expressive client-specific adaptation without additional communication. Client- and layer-specific mixing coefficients jointly fuse the two branches, and only the shared LoRA parameters are exchanged. A controlled linear study supports this role assignment: LoRA yields more aligned client updates and lower aggregation error, while RandLoRA more accurately recovers client-specific residuals. Experiments across four visual recognition benchmarks show that FedLAFP consistently outperforms local-only and federated LoRA baselines, achieving an average personalized accuracy of $86.93\%$ and exceeding the best baseline average by $1.30$ percentage points.
