paper-with-me

Papers

Personalized Collaborative Fine-Tuning for On-Device Large Language Models

2024-04-15 · Nicolas Wagner, Dongyang Fan, Martin Jaggi

We explore on-device self-supervised collaborative fine-tuning of large language models with limited local data availability. Taking inspiration from the collaborative learning community, we introduce three distinct trust-weighted gradient aggregation schemes: weight similarity-based, prediction similarity-based and validation performance-based. To minimize communication overhead, we integrate Low-Rank Adaptation (LoRA) and only exchange LoRA weight updates. Our protocols, driven by prediction and performance metrics, surpass both FedAvg and local fine-tuning methods, which is particularly evident in realistic scenarios with more diverse local data distributions. The results underscore the effectiveness of our approach in addressing heterogeneity and scarcity within local datasets.

📄 PDF Abstract BibTeX arXiv:2404.09753

Code (1)

epfml/personalized-collaborative-llms 공식 구현 pytorch

Tasks

Prediction

Similar Papers 제목 키워드 기반

Flexible Personalized Split Federated Learning for On-Device Fine-Tuning of Foundation Models

2025-08-14 · Tianjun Yuan, Jiaxiang Geng, Pengchao Han, Xianhao Chen 외 arxiv

Fine-tuning foundation models is critical for superior performance on personalized downstream tasks, compared to using pre-trained models. Collaborative learning can leverage local clients' datasets for fine-tuning, but …

Personalized Federated Learning

DUET: A Tuning-Free Device-Cloud Collaborative Parameters Generation Framework for Efficient Device Model Generalization

2022-09-12 · Zheqi Lv, Wenqiao Zhang, Shengyu Zhang, Kun Kuang 외

Device Model Generalization (DMG) is a practical yet under-investigated research topic for on-device machine learning applications. It aims to improve the generalization ability of pre-trained models when deployed on res…

Device-Cloud CollaborationDomain AdaptationImage ClassificationModel Compression+1

CoSteer: Collaborative Decoding-Time Personalization via Local Delta Steering

2025-07-07 · Hang Lv, Sheng Liang, Hao Wang, Hongchao Gu 외

Personalized text generation has become crucial for adapting language models to diverse and evolving users' personal context across cultural, temporal, and contextual dimensions. While existing methods often rely on cent…

Text Generation

Fed MobiLLM: Efficient Federated LLM Fine-Tuning over Heterogeneous Mobile Devices via Server Assisted Side-Tuning

2025-08-09 · Xingke Yang, Liang Li, Sicong Li, Liwei Guan 외 arxiv

Collaboratively fine-tuning (FT) large language models (LLMs) over heterogeneous mobile devices fosters immense potential applications of personalized intelligence. However, such a vision faces critical system challenges…

PocketLLM: Enabling On-Device Fine-Tuning for Personalized LLMs

2024-07-01 · Dan Peng, Zhihui Fu, Jun Wang

Recent advancements in large language models (LLMs) have indeed showcased their impressive capabilities. On mobile devices, the wealth of valuable, non-public data generated daily holds great promise for locally fine-tun…