Enregistré dans:
Détails bibliographiques
Auteurs principaux: Guo, Wei, Lu, Siyuan, Tong, Yiqi, Hu, Zhaojun, Zhuang, Fuzhen, Zhang, Xiao, Fan, Tao, Dong, Jin
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2507.22633
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866908473247137792
author Guo, Wei
Lu, Siyuan
Tong, Yiqi
Hu, Zhaojun
Zhuang, Fuzhen
Zhang, Xiao
Fan, Tao
Dong, Jin
author_facet Guo, Wei
Lu, Siyuan
Tong, Yiqi
Hu, Zhaojun
Zhuang, Fuzhen
Zhang, Xiao
Fan, Tao
Dong, Jin
contents Different from existing federated fine-tuning (FFT) methods for foundation models, hybrid heterogeneous federated fine-tuning (HHFFT) is an under-explored scenario where clients exhibit double heterogeneity in model architectures and downstream tasks. This hybrid heterogeneity introduces two significant challenges: 1) heterogeneous matrix aggregation, where clients adopt different large-scale foundation models based on their task requirements and resource limitations, leading to dimensional mismatches during LoRA parameter aggregation; and 2) multi-task knowledge interference, where local shared parameters, trained with both task-shared and task-specific knowledge, cannot ensure only task-shared knowledge is transferred between clients. To address these challenges, we propose H2Tune, a federated foundation model fine-tuning with hybrid heterogeneity. Our framework H2Tune consists of three key components: (i) sparsified triple matrix decomposition to align hidden dimensions across clients through constructing rank-consistent middle matrices, with adaptive sparsification based on client resources; (ii) relation-guided matrix layer alignment to handle heterogeneous layer structures and representation capabilities; and (iii) alternating task-knowledge disentanglement mechanism to decouple shared and specific knowledge of local model parameters through alternating optimization. Theoretical analysis proves a convergence rate of O(1/\sqrt{T}). Extensive experiments show our method achieves up to 15.4% accuracy improvement compared to state-of-the-art baselines. Our code is available at https://anonymous.4open.science/r/H2Tune-1407.
format Preprint
id arxiv_https___arxiv_org_abs_2507_22633
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle H2Tune: Federated Foundation Model Fine-Tuning with Hybrid Heterogeneity
Guo, Wei
Lu, Siyuan
Tong, Yiqi
Hu, Zhaojun
Zhuang, Fuzhen
Zhang, Xiao
Fan, Tao
Dong, Jin
Machine Learning
Artificial Intelligence
Different from existing federated fine-tuning (FFT) methods for foundation models, hybrid heterogeneous federated fine-tuning (HHFFT) is an under-explored scenario where clients exhibit double heterogeneity in model architectures and downstream tasks. This hybrid heterogeneity introduces two significant challenges: 1) heterogeneous matrix aggregation, where clients adopt different large-scale foundation models based on their task requirements and resource limitations, leading to dimensional mismatches during LoRA parameter aggregation; and 2) multi-task knowledge interference, where local shared parameters, trained with both task-shared and task-specific knowledge, cannot ensure only task-shared knowledge is transferred between clients. To address these challenges, we propose H2Tune, a federated foundation model fine-tuning with hybrid heterogeneity. Our framework H2Tune consists of three key components: (i) sparsified triple matrix decomposition to align hidden dimensions across clients through constructing rank-consistent middle matrices, with adaptive sparsification based on client resources; (ii) relation-guided matrix layer alignment to handle heterogeneous layer structures and representation capabilities; and (iii) alternating task-knowledge disentanglement mechanism to decouple shared and specific knowledge of local model parameters through alternating optimization. Theoretical analysis proves a convergence rate of O(1/\sqrt{T}). Extensive experiments show our method achieves up to 15.4% accuracy improvement compared to state-of-the-art baselines. Our code is available at https://anonymous.4open.science/r/H2Tune-1407.
title H2Tune: Federated Foundation Model Fine-Tuning with Hybrid Heterogeneity
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2507.22633