Null-Space Filtering for Data-Free Continual Model Merging: Preserving Stability, Promoting Plasticity

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Qiu, Zihuan, Wang, Lei, Cao, Yang, Zhang, Runtong, Su, Bing, Xu, Yi, Meng, Fanman, Xu, Linfeng, Wu, Qingbo, Li, Hongliang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915863635951616
author Qiu, Zihuan
Wang, Lei
Cao, Yang
Zhang, Runtong
Su, Bing
Xu, Yi
Meng, Fanman
Xu, Linfeng
Wu, Qingbo
Li, Hongliang
author_facet Qiu, Zihuan
Wang, Lei
Cao, Yang
Zhang, Runtong
Su, Bing
Xu, Yi
Meng, Fanman
Xu, Linfeng
Wu, Qingbo
Li, Hongliang
contents Data-free continual model merging (DFCMM) aims to fuse independently fine-tuned models into a single backbone that evolves with incoming tasks without accessing task data. This paper revisits two fundamental desiderata for DFCMM: stability, avoiding interference with earlier tasks, and plasticity, adapting faithfully to each new task. This poses a challenge that existing approaches fail to address: how to bridge data-level desiderata with parameter-space optimization to ensure stability and plasticity in the absence of task data. To this end, we propose NUFILT (NUll-space FILTering), a data-free framework that directly links these desiderata into parameter-space optimization. Our key observation is that task vectors approximately align with representation subspaces, providing structural surrogates for enforcing stability and plasticity. Accordingly, we design a null-space projector that preserves prior responses by filtering overlapping components of new task vectors, ensuring stability. We further introduce a lightweight LoRA adapter that injects complementary task-specific signals to enable plasticity. The adapter is trained with a projection-based surrogate loss that preserves consistency with prior knowledge while introducing novel directions. This joint filtering-adaptation process enables the backbone to absorb new knowledge while retaining existing behaviors, with updates fused back in a layer-wise linear fashion without extra parameters or inference cost. Theoretically, we establish approximate subspace alignment guarantees that justify null-space filtering. Empirically, NUFILT achieves state-of-the-art performance with minimal forgetting on both vision and NLP benchmarks, improving average accuracy by 4-7% over OPCM and WUDI-Merging, while narrowing the gap to fine-tuning and reducing computation overhead. The code is available at: https://github.com/zihuanqiu/NUFILT
format Preprint
id arxiv_https___arxiv_org_abs_2509_21413
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Null-Space Filtering for Data-Free Continual Model Merging: Preserving Stability, Promoting Plasticity
Qiu, Zihuan
Wang, Lei
Cao, Yang
Zhang, Runtong
Su, Bing
Xu, Yi
Meng, Fanman
Xu, Linfeng
Wu, Qingbo
Li, Hongliang
Machine Learning
Data-free continual model merging (DFCMM) aims to fuse independently fine-tuned models into a single backbone that evolves with incoming tasks without accessing task data. This paper revisits two fundamental desiderata for DFCMM: stability, avoiding interference with earlier tasks, and plasticity, adapting faithfully to each new task. This poses a challenge that existing approaches fail to address: how to bridge data-level desiderata with parameter-space optimization to ensure stability and plasticity in the absence of task data. To this end, we propose NUFILT (NUll-space FILTering), a data-free framework that directly links these desiderata into parameter-space optimization. Our key observation is that task vectors approximately align with representation subspaces, providing structural surrogates for enforcing stability and plasticity. Accordingly, we design a null-space projector that preserves prior responses by filtering overlapping components of new task vectors, ensuring stability. We further introduce a lightweight LoRA adapter that injects complementary task-specific signals to enable plasticity. The adapter is trained with a projection-based surrogate loss that preserves consistency with prior knowledge while introducing novel directions. This joint filtering-adaptation process enables the backbone to absorb new knowledge while retaining existing behaviors, with updates fused back in a layer-wise linear fashion without extra parameters or inference cost. Theoretically, we establish approximate subspace alignment guarantees that justify null-space filtering. Empirically, NUFILT achieves state-of-the-art performance with minimal forgetting on both vision and NLP benchmarks, improving average accuracy by 4-7% over OPCM and WUDI-Merging, while narrowing the gap to fine-tuning and reducing computation overhead. The code is available at: https://github.com/zihuanqiu/NUFILT
title Null-Space Filtering for Data-Free Continual Model Merging: Preserving Stability, Promoting Plasticity
topic Machine Learning
url https://arxiv.org/abs/2509.21413