Integrating Task-Specific and Universal Adapters for Pre-Trained Model-based Class-Incremental Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Yan, Zhou, Da-Wei, Ye, Han-Jia
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918122203643904
author Wang, Yan
Zhou, Da-Wei
Ye, Han-Jia
author_facet Wang, Yan
Zhou, Da-Wei
Ye, Han-Jia
contents Class-Incremental Learning (CIL) requires a learning system to continually learn new classes without forgetting. Existing pre-trained model-based CIL methods often freeze the pre-trained network and adapt to incremental tasks using additional lightweight modules such as adapters. However, incorrect module selection during inference hurts performance, and task-specific modules often overlook shared general knowledge, leading to errors on distinguishing between similar classes across tasks. To address the aforementioned challenges, we propose integrating Task-Specific and Universal Adapters (TUNA) in this paper. Specifically, we train task-specific adapters to capture the most crucial features relevant to their respective tasks and introduce an entropy-based selection mechanism to choose the most suitable adapter. Furthermore, we leverage an adapter fusion strategy to construct a universal adapter, which encodes the most discriminative features shared across tasks. We combine task-specific and universal adapter predictions to harness both specialized and general knowledge during inference. Extensive experiments on various benchmark datasets demonstrate the state-of-the-art performance of our approach. Code is available at: https://github.com/LAMDA-CL/ICCV2025-TUNA
format Preprint
id arxiv_https___arxiv_org_abs_2508_08165
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Integrating Task-Specific and Universal Adapters for Pre-Trained Model-based Class-Incremental Learning
Wang, Yan
Zhou, Da-Wei
Ye, Han-Jia
Computer Vision and Pattern Recognition
Machine Learning
Class-Incremental Learning (CIL) requires a learning system to continually learn new classes without forgetting. Existing pre-trained model-based CIL methods often freeze the pre-trained network and adapt to incremental tasks using additional lightweight modules such as adapters. However, incorrect module selection during inference hurts performance, and task-specific modules often overlook shared general knowledge, leading to errors on distinguishing between similar classes across tasks. To address the aforementioned challenges, we propose integrating Task-Specific and Universal Adapters (TUNA) in this paper. Specifically, we train task-specific adapters to capture the most crucial features relevant to their respective tasks and introduce an entropy-based selection mechanism to choose the most suitable adapter. Furthermore, we leverage an adapter fusion strategy to construct a universal adapter, which encodes the most discriminative features shared across tasks. We combine task-specific and universal adapter predictions to harness both specialized and general knowledge during inference. Extensive experiments on various benchmark datasets demonstrate the state-of-the-art performance of our approach. Code is available at: https://github.com/LAMDA-CL/ICCV2025-TUNA
title Integrating Task-Specific and Universal Adapters for Pre-Trained Model-based Class-Incremental Learning
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2508.08165