Ravan: Multi-Head Low-Rank Adaptation for Federated Fine-Tuning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Raje, Arian, Askin, Baris, Jhunjhunwala, Divyansh, Joshi, Gauri
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912862312595456
author Raje, Arian
Askin, Baris
Jhunjhunwala, Divyansh
Joshi, Gauri
author_facet Raje, Arian
Askin, Baris
Jhunjhunwala, Divyansh
Joshi, Gauri
contents Large language models (LLMs) have not yet effectively leveraged the vast amounts of edge-device data, and federated learning (FL) offers a promising paradigm to collaboratively fine-tune LLMs without transferring private edge data to the cloud. To operate within the computation and communication constraints of edge devices, recent literature on federated fine-tuning of LLMs proposes the use of low-rank adaptation (LoRA) and similar parameter-efficient methods. However, LoRA-based methods suffer from accuracy degradation in FL settings, primarily because of data and computational heterogeneity across clients. We propose Ravan, an adaptive multi-head LoRA method that balances parameter efficiency and model expressivity by reparameterizing the weight updates as the sum of multiple LoRA heads $s_i\textbf{B}_i\textbf{H}_i\textbf{A}_i$ in which only the core matrices $\textbf{H}_i$ and their lightweight scaling factors $s_i$ are trained. These trainable scaling factors let the optimization focus on the most useful heads, recovering a higher-rank approximation of the full update without increasing the number of communicated parameters since clients upload $s_i\textbf{H}_i$ directly. Experiments on vision and language benchmarks show that Ravan improves test accuracy by $2-8\%$ over prior parameter-efficient baselines, making it a robust and scalable solution for federated fine-tuning of LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2506_05568
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Ravan: Multi-Head Low-Rank Adaptation for Federated Fine-Tuning
Raje, Arian
Askin, Baris
Jhunjhunwala, Divyansh
Joshi, Gauri
Machine Learning
Artificial Intelligence
Large language models (LLMs) have not yet effectively leveraged the vast amounts of edge-device data, and federated learning (FL) offers a promising paradigm to collaboratively fine-tune LLMs without transferring private edge data to the cloud. To operate within the computation and communication constraints of edge devices, recent literature on federated fine-tuning of LLMs proposes the use of low-rank adaptation (LoRA) and similar parameter-efficient methods. However, LoRA-based methods suffer from accuracy degradation in FL settings, primarily because of data and computational heterogeneity across clients. We propose Ravan, an adaptive multi-head LoRA method that balances parameter efficiency and model expressivity by reparameterizing the weight updates as the sum of multiple LoRA heads $s_i\textbf{B}_i\textbf{H}_i\textbf{A}_i$ in which only the core matrices $\textbf{H}_i$ and their lightweight scaling factors $s_i$ are trained. These trainable scaling factors let the optimization focus on the most useful heads, recovering a higher-rank approximation of the full update without increasing the number of communicated parameters since clients upload $s_i\textbf{H}_i$ directly. Experiments on vision and language benchmarks show that Ravan improves test accuracy by $2-8\%$ over prior parameter-efficient baselines, making it a robust and scalable solution for federated fine-tuning of LLMs.
title Ravan: Multi-Head Low-Rank Adaptation for Federated Fine-Tuning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.05568