Two Stage Wireless Federated LoRA Fine-Tuning with Sparsified Orthogonal Updates

TL;DR

This paper proposes a novel wireless federated LoRA fine-tuning framework with adaptive sparsification and a two-stage algorithm, significantly reducing communication costs while maintaining high accuracy in large language model training.

Contribution

It introduces SOFT, an adaptive sparsification method, and TSFA, a two-stage algorithm for efficient wireless federated LoRA fine-tuning, with theoretical convergence analysis.

Findings

Achieves comparable accuracy to ideal models

Reduces communication overhead significantly

Enables scalable large model deployment in wireless FL

Abstract

Transformer-based large language models (LLMs) have achieved remarkable success across various tasks. Yet, fine-tuning such massive models in federated learning (FL) settings poses significant challenges due to resource constraints and communication overhead. Low-Rank Adaptation (LoRA) addresses these issues by training compact, low-rank matrices instead of fully fine-tuning large models. This paper introduces a wireless federated LoRA fine-tuning framework that optimizes both learning performance and communication efficiency. We provide a novel convergence analysis, revealing how LoRA rank and covariance effects influence FL training dynamics. Leveraging these insights, we propose Sparsified Orthogonal Fine-Tuning (\textbf{SOFT}), an adaptive sparsification method that streamlines parameter updates without expensive matrix multiplications and singular value decomposition (SVD) operations. Additionally, we present a Two Stage Federated Algorithm (\textbf{TSFA}) algorithm that pre-determines key parameters offline and dynamically adjusts bandwidth and sparsification online, ensuring efficient training under latency constraints. Experiments on benchmark datasets show that our approach achieves accuracy comparable to ideal scenario models while significantly reducing communication overhead. Our framework thus enables scalable, resource-efficient deployment of large models in real-world wireless FL scenarios.