paper-with-me

홈 › Papers

ASLoRA: Adaptive Sharing Low-Rank Adaptation Across Layers

2024-12-13 · Junyan Hu, Xue Xiao, Mengqi Zhang, Yao Chen, Zhaochun Ren, Zhumin Chen, Pengjie Ren

As large language models (LLMs) grow in size, traditional full fine-tuning becomes increasingly impractical due to its high computational and storage costs. Although popular parameter-efficient fine-tuning methods, such as LoRA, have significantly reduced the number of tunable parameters, there is still room for further optimization. In this work, we propose ASLoRA, a cross-layer parameter-sharing strategy combining global sharing with partial adaptive sharing. Specifically, we share the low-rank matrix A across all layers and adaptively merge matrix B during training. This sharing mechanism not only mitigates overfitting effectively but also captures inter-layer dependencies, significantly enhancing the model's representational capability. We conduct extensive experiments on various NLP tasks, showing that ASLoRA outperforms LoRA while using less than 25% of the parameters, highlighting its flexibility and superior parameter efficiency. Furthermore, in-depth analyses of the adaptive sharing strategy confirm its significant advantages in enhancing both model flexibility and task adaptability.

📄 PDF Abstract BibTeX arXiv:2412.10135

Code (0)

등록된 구현이 없습니다.

Tasks

parameter-efficient fine-tuning

Similar Papers 제목 키워드 기반

Adaptive Shared Experts with LoRA-Based Mixture of Experts for Multi-Task Learning

2025-10-01 · Minghao Yang, Ren Togo, Guang Li, Takahiro Ogawa 외 arxiv

Mixture-of-Experts (MoE) has emerged as a powerful framework for multi-task learning (MTL). However, existing MoE-MTL methods often rely on single-task pretrained backbones and suffer from redundant adaptation and ineffi…

Multi-Task Learning

DiffoRA: Enabling Parameter-Efficient LLM Fine-Tuning via Differential Low-Rank Matrix Adaptation

2025-02-13 · Tangyu Jiang, Haodi Wang, Chun Yuan

The Parameter-Efficient Fine-Tuning (PEFT) methods have been extensively researched for large language models in the downstream tasks. Among all the existing approaches, the Low-Rank Adaptation (LoRA) has gained populari…

parameter-efficient fine-tuning

HiLoRA: Hierarchical Low-Rank Adaptation for Personalized Federated Learning

2026-03-03 · Zihao Peng, Nan Zou, Jiandian Zeng, Guo Li 외 arxiv

Vision Transformers (ViTs) have been widely adopted in vision tasks due to their strong transferability. In Federated Learning (FL), where full fine-tuning is communication heavy, Low-Rank Adaptation (LoRA) provides an e…

Personalized Federated LearningRepresentation Learning

FedCTTA: A Collaborative Approach to Continual Test-Time Adaptation in Federated Learning

2025-05-19 · Rakibul Hasan Rajib, Md Akil Raihan Iftee, Mir Sazzat Hossain, A. K. M. Mahbubur Rahman 외

Federated Learning (FL) enables collaborative model training across distributed clients without sharing raw data, making it ideal for privacy-sensitive applications. However, FL models often suffer performance degradatio…

Federated LearningPrivacy PreservingTest-time Adaptation

Hypernetwork-Driven Low-Rank Adaptation Across Attention Heads

2025-10-05 · Nghiem T. Diep, Dung Le, Tuan Truong, Tan Dinh 외 arxiv

Parameter-efficient fine-tuning (PEFT) has emerged as a powerful paradigm for adapting large-scale pre-trained models to downstream tasks with minimal additional parameters. Among PEFT methods, Low-Rank Adaptation (LoRA)…

parameter-efficient fine-tuning