CALM: Consensus-Aware Localized Merging for Multi-Task Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yan, Kunda, Zhang, Min, Cui, Sen, Qu, Zikun, Jiang, Bo, Liu, Feng, Zhang, Changshui
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909650135285760
author Yan, Kunda
Zhang, Min
Cui, Sen
Qu, Zikun
Jiang, Bo
Liu, Feng
Zhang, Changshui
author_facet Yan, Kunda
Zhang, Min
Cui, Sen
Qu, Zikun
Jiang, Bo
Liu, Feng
Zhang, Changshui
contents Model merging aims to integrate the strengths of multiple fine-tuned models into a unified model while preserving task-specific capabilities. Existing methods, represented by task arithmetic, are typically classified into global- and local-aware methods. However, global-aware methods inevitably cause parameter interference, while local-aware methods struggle to maintain the effectiveness of task-specific details in the merged model. To address these limitations, we propose a Consensus-Aware Localized Merging (CALM) method which incorporates localized information aligned with global task consensus, ensuring its effectiveness post-merging. CALM consists of three key components: (1) class-balanced entropy minimization sampling, providing a more flexible and reliable way to leverage unsupervised data; (2) an efficient-aware framework, selecting a small set of tasks for sequential merging with high scalability; (3) a consensus-aware mask optimization, aligning localized binary masks with global task consensus and merging them conflict-free. Experiments demonstrate the superiority and robustness of our CALM, significantly outperforming existing methods and achieving performance close to traditional MTL.
format Preprint
id arxiv_https___arxiv_org_abs_2506_13406
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CALM: Consensus-Aware Localized Merging for Multi-Task Learning
Yan, Kunda
Zhang, Min
Cui, Sen
Qu, Zikun
Jiang, Bo
Liu, Feng
Zhang, Changshui
Machine Learning
Artificial Intelligence
Model merging aims to integrate the strengths of multiple fine-tuned models into a unified model while preserving task-specific capabilities. Existing methods, represented by task arithmetic, are typically classified into global- and local-aware methods. However, global-aware methods inevitably cause parameter interference, while local-aware methods struggle to maintain the effectiveness of task-specific details in the merged model. To address these limitations, we propose a Consensus-Aware Localized Merging (CALM) method which incorporates localized information aligned with global task consensus, ensuring its effectiveness post-merging. CALM consists of three key components: (1) class-balanced entropy minimization sampling, providing a more flexible and reliable way to leverage unsupervised data; (2) an efficient-aware framework, selecting a small set of tasks for sequential merging with high scalability; (3) a consensus-aware mask optimization, aligning localized binary masks with global task consensus and merging them conflict-free. Experiments demonstrate the superiority and robustness of our CALM, significantly outperforming existing methods and achieving performance close to traditional MTL.
title CALM: Consensus-Aware Localized Merging for Multi-Task Learning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.13406