Fine-tuning large language models (LLMs) on private, on-device data can empower tailored personalized AI agents. However, fine-tuning LLMs on resource-constrained edge devices faces significant challenges, including excessive computation overhead, device heterogeneity, and data imbalance. This paper proposes SplitFrozen, a split learning framework that enables efficient LLM fine-tuning by strategically freezing device-side model layers while centralizing parameter-efficient fine-tuning on the server. Our framework partitions LLMs into device-side frozen layers and server-side fine-tuning layers, where heterogeneous resource-constrained devices execute only forward propagation. To minimize server-side training costs, we integrate Low-Rank Adaptation (LoRA) into the server-side layers. A pipeline parallelism strategy further optimizes training efficiency by decoupling device-server computations and leveraging decomposed backward propagation. Experiments on GPT-2 with the MRPC, MNLI-matched, and SST-2 datasets demonstrate that SplitFrozen outperforms FedLoRA and SplitLoRA by 69.4\% model accuracy under extremely imbalanced data, while reducing up to 86.8\% device-side computations and 50.2\% total training time. Experiments also validate the scalability of SplitFrozen on content generation task using Llama-3.2 model on GSM8K dataset.
View on arXiv@article{ma2025_2503.18986,
title={ SplitFrozen: Split Learning with Device-side Model Frozen for Fine-Tuning LLM on Heterogeneous Resource-Constrained Devices },
author={ Jian Ma and Xinchen Lyu and Jun Jiang and Qimei Cui and Haipeng Yao and Xiaofeng Tao },
journal={arXiv preprint arXiv:2503.18986},
year={ 2025 }
}