A Simple Efficiency Incremental Learning Framework via Vision-Language Model with Nonlinear Multi-Adapters
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910058712924160 |
|---|---|
| author | Luo, Haihua Ran, Xuming Shen, Jiangrong Hämäläinen, Timo Chen, Zhonghua Xu, Qi Cong, Fengyu |
| author_facet | Luo, Haihua Ran, Xuming Shen, Jiangrong Hämäläinen, Timo Chen, Zhonghua Xu, Qi Cong, Fengyu |
| contents | Incremental Learning (IL) aims to learn new tasks while preserving previously acquired knowledge. Integrating the zero-shot learning capabilities of pre-trained vision-language models into IL methods has marked a significant advancement. However, these methods face three primary challenges: (1) the need for improved training efficiency; (2) reliance on a memory bank to store previous data; and (3) the necessity of a strong backbone to augment the model's capabilities. In this paper, we propose SimE, a Simple and Efficient framework that employs a vision-language model with adapters designed specifically for the IL task. We report a remarkable phenomenon: there is a nonlinear correlation between the number of adaptive adapter connections and the model's IL capabilities. While increasing adapter connections between transformer blocks improves model performance, adding more adaptive connections within transformer blocks during smaller incremental steps does not enhance, and may even degrade the model's IL ability. Extensive experimental results show that SimE surpasses traditional methods by 9.6% on TinyImageNet and outperforms other CLIP-based methods by 5.3% on CIFAR-100. Furthermore, we conduct a systematic study to enhance the utilization of the zero-shot capabilities of CLIP. We suggest replacing SimE's encoder with a CLIP model trained on larger datasets (e.g., LAION2B) and stronger architectures (e.g., ViT-L/14). |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_11211 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | A Simple Efficiency Incremental Learning Framework via Vision-Language Model with Nonlinear Multi-Adapters Luo, Haihua Ran, Xuming Shen, Jiangrong Hämäläinen, Timo Chen, Zhonghua Xu, Qi Cong, Fengyu Computer Vision and Pattern Recognition Artificial Intelligence Incremental Learning (IL) aims to learn new tasks while preserving previously acquired knowledge. Integrating the zero-shot learning capabilities of pre-trained vision-language models into IL methods has marked a significant advancement. However, these methods face three primary challenges: (1) the need for improved training efficiency; (2) reliance on a memory bank to store previous data; and (3) the necessity of a strong backbone to augment the model's capabilities. In this paper, we propose SimE, a Simple and Efficient framework that employs a vision-language model with adapters designed specifically for the IL task. We report a remarkable phenomenon: there is a nonlinear correlation between the number of adaptive adapter connections and the model's IL capabilities. While increasing adapter connections between transformer blocks improves model performance, adding more adaptive connections within transformer blocks during smaller incremental steps does not enhance, and may even degrade the model's IL ability. Extensive experimental results show that SimE surpasses traditional methods by 9.6% on TinyImageNet and outperforms other CLIP-based methods by 5.3% on CIFAR-100. Furthermore, we conduct a systematic study to enhance the utilization of the zero-shot capabilities of CLIP. We suggest replacing SimE's encoder with a CLIP model trained on larger datasets (e.g., LAION2B) and stronger architectures (e.g., ViT-L/14). |
| title | A Simple Efficiency Incremental Learning Framework via Vision-Language Model with Nonlinear Multi-Adapters |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2603.11211 |