Neural Parameter Search for Slimmer Fine-Tuned Models and Better Transfer

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Du, Guodong, Fang, Zitao, Li, Jing, Li, Junlin, Jiang, Runhua, Yu, Shuyang, Guo, Yifei, Chen, Yangneng, Goh, Sim Kuan, Tang, Ho-Kin, He, Daojing, Liu, Honghai, Zhang, Min
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915303487700992
author Du, Guodong
Fang, Zitao
Li, Jing
Li, Junlin
Jiang, Runhua
Yu, Shuyang
Guo, Yifei
Chen, Yangneng
Goh, Sim Kuan
Tang, Ho-Kin
He, Daojing
Liu, Honghai
Zhang, Min
author_facet Du, Guodong
Fang, Zitao
Li, Jing
Li, Junlin
Jiang, Runhua
Yu, Shuyang
Guo, Yifei
Chen, Yangneng
Goh, Sim Kuan
Tang, Ho-Kin
He, Daojing
Liu, Honghai
Zhang, Min
contents Foundation models and their checkpoints have significantly advanced deep learning, boosting performance across various applications. However, fine-tuned models often struggle outside their specific domains and exhibit considerable redundancy. Recent studies suggest that combining a pruned fine-tuned model with the original pre-trained model can mitigate forgetting, reduce interference when merging model parameters across tasks, and improve compression efficiency. In this context, developing an effective pruning strategy for fine-tuned models is crucial. Leveraging the advantages of the task vector mechanism, we preprocess fine-tuned models by calculating the differences between them and the original model. Recognizing that different task vector subspaces contribute variably to model performance, we introduce a novel method called Neural Parameter Search (NPS-Pruning) for slimming down fine-tuned models. This method enhances pruning efficiency by searching through neural parameters of task vectors within low-rank subspaces. Our method has three key applications: enhancing knowledge transfer through pairwise model interpolation, facilitating effective knowledge fusion via model merging, and enabling the deployment of compressed models that retain near-original performance while significantly reducing storage costs. Extensive experiments across vision, NLP, and multi-modal benchmarks demonstrate the effectiveness and robustness of our approach, resulting in substantial performance gains. The code is publicly available at: https://github.com/duguodong7/NPS-Pruning.
format Preprint
id arxiv_https___arxiv_org_abs_2505_18713
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Neural Parameter Search for Slimmer Fine-Tuned Models and Better Transfer
Du, Guodong
Fang, Zitao
Li, Jing
Li, Junlin
Jiang, Runhua
Yu, Shuyang
Guo, Yifei
Chen, Yangneng
Goh, Sim Kuan
Tang, Ho-Kin
He, Daojing
Liu, Honghai
Zhang, Min
Machine Learning
Artificial Intelligence
Computation and Language
Foundation models and their checkpoints have significantly advanced deep learning, boosting performance across various applications. However, fine-tuned models often struggle outside their specific domains and exhibit considerable redundancy. Recent studies suggest that combining a pruned fine-tuned model with the original pre-trained model can mitigate forgetting, reduce interference when merging model parameters across tasks, and improve compression efficiency. In this context, developing an effective pruning strategy for fine-tuned models is crucial. Leveraging the advantages of the task vector mechanism, we preprocess fine-tuned models by calculating the differences between them and the original model. Recognizing that different task vector subspaces contribute variably to model performance, we introduce a novel method called Neural Parameter Search (NPS-Pruning) for slimming down fine-tuned models. This method enhances pruning efficiency by searching through neural parameters of task vectors within low-rank subspaces. Our method has three key applications: enhancing knowledge transfer through pairwise model interpolation, facilitating effective knowledge fusion via model merging, and enabling the deployment of compressed models that retain near-original performance while significantly reducing storage costs. Extensive experiments across vision, NLP, and multi-modal benchmarks demonstrate the effectiveness and robustness of our approach, resulting in substantial performance gains. The code is publicly available at: https://github.com/duguodong7/NPS-Pruning.
title Neural Parameter Search for Slimmer Fine-Tuned Models and Better Transfer
topic Machine Learning
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2505.18713