PrunePEFT: Iterative Hybrid Pruning for Parameter-Efficient Fine-tuning of LLMs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yu, Tongzhou, Zhang, Zhuhao, Zhu, Guanghui, Jiang, Shen, Qiu, Meikang, Huang, Yihua
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913885192192000
author Yu, Tongzhou
Zhang, Zhuhao
Zhu, Guanghui
Jiang, Shen
Qiu, Meikang
Huang, Yihua
author_facet Yu, Tongzhou
Zhang, Zhuhao
Zhu, Guanghui
Jiang, Shen
Qiu, Meikang
Huang, Yihua
contents Parameter Efficient Fine-Tuning (PEFT) methods have emerged as effective and promising approaches for fine-tuning pre-trained language models. Compared with Full parameter Fine-Tuning (FFT), PEFT achieved comparable task performance with a substantial reduction of trainable parameters, which largely saved the training and storage costs. However, using the PEFT method requires considering a vast design space, such as the type of PEFT modules and their insertion layers. Inadequate configurations can lead to sub-optimal results. Conventional solutions such as architectural search techniques, while effective, tend to introduce substantial additional overhead. In this paper, we propose a novel approach, PrunePEFT, which formulates the PEFT strategy search as a pruning problem and introduces a hybrid pruning strategy that capitalizes on the sensitivity of pruning methods to different PEFT modules. This method extends traditional pruning techniques by iteratively removing redundant or conflicting PEFT modules, thereby optimizing the fine-tuned configuration. By efficiently identifying the most relevant modules, our approach significantly reduces the computational burden typically associated with architectural search processes, making it a more scalable and efficient solution for fine-tuning large pre-trained models.
format Preprint
id arxiv_https___arxiv_org_abs_2506_07587
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PrunePEFT: Iterative Hybrid Pruning for Parameter-Efficient Fine-tuning of LLMs
Yu, Tongzhou
Zhang, Zhuhao
Zhu, Guanghui
Jiang, Shen
Qiu, Meikang
Huang, Yihua
Machine Learning
Artificial Intelligence
Parameter Efficient Fine-Tuning (PEFT) methods have emerged as effective and promising approaches for fine-tuning pre-trained language models. Compared with Full parameter Fine-Tuning (FFT), PEFT achieved comparable task performance with a substantial reduction of trainable parameters, which largely saved the training and storage costs. However, using the PEFT method requires considering a vast design space, such as the type of PEFT modules and their insertion layers. Inadequate configurations can lead to sub-optimal results. Conventional solutions such as architectural search techniques, while effective, tend to introduce substantial additional overhead. In this paper, we propose a novel approach, PrunePEFT, which formulates the PEFT strategy search as a pruning problem and introduces a hybrid pruning strategy that capitalizes on the sensitivity of pruning methods to different PEFT modules. This method extends traditional pruning techniques by iteratively removing redundant or conflicting PEFT modules, thereby optimizing the fine-tuned configuration. By efficiently identifying the most relevant modules, our approach significantly reduces the computational burden typically associated with architectural search processes, making it a more scalable and efficient solution for fine-tuning large pre-trained models.
title PrunePEFT: Iterative Hybrid Pruning for Parameter-Efficient Fine-tuning of LLMs
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.07587