P2L-CA: An Effective Parameter Tuning Framework for Rehearsal-Free Multi-Label Class-Incremental Learning

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Dong, Songlin, Li, Jiangyang, Ding, Chenhao, Ma, Zhiheng, Luo, Haoyu, He, Yuhang, Gong, Yihong
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918295136894976
author Dong, Songlin
Li, Jiangyang
Ding, Chenhao
Ma, Zhiheng
Luo, Haoyu
He, Yuhang
Gong, Yihong
author_facet Dong, Songlin
Li, Jiangyang
Ding, Chenhao
Ma, Zhiheng
Luo, Haoyu
He, Yuhang
Gong, Yihong
contents Multi-label Class-Incremental Learning aims to continuously recognize novel categories in complex scenes where multiple objects co-occur. However, existing approaches often incur high computational costs due to full-parameter fine-tuning and substantial storage overhead from memory buffers, or they struggle to address feature confusion and domain discrepancies adequately. To overcome these limitations, we introduce P2L-CA, a parameter-efficient framework that integrates a Prompt-to-Label module with a Continuous Adapter module. The P2L module leverages class-specific prompts to disentangle multi-label representations while incorporating linguistic priors to enforce stable semantic-visual alignment. Meanwhile, the CA module employs lightweight adapters to mitigate domain gaps between pre-trained models and downstream tasks, thereby enhancing model plasticity. Extensive experiments across standard and challenging MLCIL settings on MS-COCO and PASCAL VOC show that P2L-CA not only achieves substantial improvements over state-of-the-art methods but also demonstrates strong generalization in CIL scenarios, all while requiring minimal trainable parameters and eliminating the need for memory buffers.
format Preprint
id arxiv_https___arxiv_org_abs_2601_12714
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle P2L-CA: An Effective Parameter Tuning Framework for Rehearsal-Free Multi-Label Class-Incremental Learning
Dong, Songlin
Li, Jiangyang
Ding, Chenhao
Ma, Zhiheng
Luo, Haoyu
He, Yuhang
Gong, Yihong
Computer Vision and Pattern Recognition
Machine Learning
Multi-label Class-Incremental Learning aims to continuously recognize novel categories in complex scenes where multiple objects co-occur. However, existing approaches often incur high computational costs due to full-parameter fine-tuning and substantial storage overhead from memory buffers, or they struggle to address feature confusion and domain discrepancies adequately. To overcome these limitations, we introduce P2L-CA, a parameter-efficient framework that integrates a Prompt-to-Label module with a Continuous Adapter module. The P2L module leverages class-specific prompts to disentangle multi-label representations while incorporating linguistic priors to enforce stable semantic-visual alignment. Meanwhile, the CA module employs lightweight adapters to mitigate domain gaps between pre-trained models and downstream tasks, thereby enhancing model plasticity. Extensive experiments across standard and challenging MLCIL settings on MS-COCO and PASCAL VOC show that P2L-CA not only achieves substantial improvements over state-of-the-art methods but also demonstrates strong generalization in CIL scenarios, all while requiring minimal trainable parameters and eliminating the need for memory buffers.
title P2L-CA: An Effective Parameter Tuning Framework for Rehearsal-Free Multi-Label Class-Incremental Learning
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2601.12714