FedQuad: Adaptive Layer-wise LoRA Deployment and Activation Quantization for Federated Fine-Tuning

TL;DR

FedQuad introduces an adaptive, resource-aware federated fine-tuning framework for large language models, combining layer-wise LoRA adjustments and activation quantization to enhance efficiency on heterogeneous, resource-limited devices.

Contribution

It proposes a novel adaptive LoRA deployment and activation quantization method that optimizes federated fine-tuning for resource-constrained and heterogeneous devices.

Findings

Achieves 1.4-5.3x faster convergence than baselines.

Effectively reduces memory and computational requirements.

Improves deployment efficiency in resource-limited environments.

Abstract

Federated fine-tuning (FedFT) provides an effective paradigm for fine-tuning large language models (LLMs) in privacy-sensitive scenarios. However, practical deployment remains challenging due to the limited resources on end devices. Existing methods typically utilize parameter-efficient fine-tuning (PEFT) techniques, such as Low-Rank Adaptation (LoRA), to substantially reduce communication overhead. Nevertheless, significant memory usage for activation storage and computational demands from full backpropagation remain major barriers to efficient deployment on resource-constrained end devices. Moreover, substantial resource heterogeneity across devices results in severe synchronization bottlenecks, diminishing the overall fine-tuning efficiency. To address these issues, we propose FedQuad, a novel LoRA-based FedFT framework that adaptively adjusts the LoRA depth (the number of consecutive tunable LoRA layers from the output) according to device computational capabilities, while employing activation quantization to reduce memory overhead, thereby enabling efficient deployment on resource-constrained devices. Specifically, FedQuad first identifies the feasible and efficient combinations of LoRA depth and the number of activation quantization layers based on device-specific resource constraints. Subsequently, FedQuad employs a greedy strategy to select the optimal configurations for each device, effectively accommodating system heterogeneity. Extensive experiments demonstrate that FedQuad achieves a 1.4-5.3x convergence acceleration compared to state-of-the-art baselines when reaching target accuracy, highlighting its efficiency and deployability in resource-constrained and heterogeneous end-device environments.