Beyond End-to-End: Dynamic Chain Optimization for Private LLM Adaptation on the Edge

TL;DR

This paper introduces ChainFed, a layer-by-layer federated fine-tuning method for private LLM adaptation on edge devices, overcoming memory limitations and improving accuracy significantly.

Contribution

It proposes a novel sequential fine-tuning paradigm with three key techniques, outperforming existing methods in privacy-preserving edge LLM adaptation.

Findings

ChainFed achieves up to 46.46% accuracy improvement.

The method effectively addresses memory constraints on edge devices.

Extensive experiments validate its superiority over existing approaches.

Abstract

Federated fine-tuning enables privacy-preserving LLM adaptation but faces a critical bottleneck: the disparity between LLMs' high memory demands and edge devices' limited capacity. To break the memory barrier, we propose Chain Federated Fine-Tuning (ChainFed), an innovative paradigm that forgoes end-to-end updates in favor of a sequential, layer-by-layer manner. It first trains the initial adapter to convergence, freezes its weights, and then proceeds to the next. This iterative train-and-freeze process forms an optimization chain, gradually enhancing the model's task-specific proficiency. ChainFed further integrates three core techniques: 1) Dynamic Layer Co-Tuning to bridge semantic gaps between sequentially tuned layers and facilitate information flow; 2) Globally Perceptive Optimization to endow each adapter with foresight beyond its local objective; 3) Function-Oriented Adaptive Tuning to automatically identify the optimal fine-tuning starting point. Extensive experiments on multiple benchmarks demonstrate the superiority of ChainFed over existing methods, boosting average accuracy by up to 46.46\%.