解决多任务联邦学习中任务异构问题，统一向量实现高效协作。
Many-Task Federated Fine-Tuning via Unified Task Vectors
- 用统一任务向量整合各客户端的多任务特征，无需分组或存储单独模型。
- 在30个数据集上表现优于现有方法，接近单任务微调效果。
- 轻量级调制模块支持相关任务知识迁移，不同任务间有效解耦。
联邦学习传统上假设客户端任务同质，但现实场景中客户端常专注不同任务，存在任务异构性。为此，多任务联邦学习（MaT-FL）应运而生，使客户端在任务多样下仍可协作。现有方法依赖客户端分组或个性化层，需服务器管理独立模型，且无法处理客户端同时承担多个任务的情况。本文提出MaTU，一种新的MaT-FL方法，通过联合学习客户端任务向量，消除对聚类和服务器端客户端专属权重存储的需求。该方法引入新颖聚合机制，基于任务向量方向判断任务相似性，并构建封装所有任务的统一任务向量。为满足任务特定需求，我们在统一向量中加入轻量级调制器，促进相关任务知识迁移，同时解耦不相似任务。在30个数据集上的评估显示，MaTU性能优于当前最优的MaT-FL方法，结果接近每任务微调水平，同时实现显著通信节省。
原文摘要 · Abstract (English)
Federated Learning (FL) traditionally assumes homogeneous client tasks; however, in real-world scenarios, clients often specialize in diverse tasks, introducing task heterogeneity. To address this challenge, Many-Task FL (MaT-FL) has emerged, enabling clients to collaborate effectively despite task diversity. Existing MaT-FL approaches rely on client grouping or personalized layers, requiring the server to manage individual models and failing to account for clients handling multiple tasks. We propose MaTU, a MaT-FL approach that enables joint learning of task vectors across clients, eliminating the need for clustering or client-specific weight storage at the server. Our method introduces a novel aggregation mechanism that determines task similarity based on the direction of clients task vectors and constructs a unified task vector encapsulating all tasks. To address task-specific requirements, we augment the unified task vector with lightweight modulators that facilitate knowledge transfer among related tasks while disentangling dissimilar ones. Evaluated across 30 datasets, MaTU achieves superior performance over state-of-the-art MaT-FL approaches, with results comparable to per-task fine-tuning, while delivering significant communication savings.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。