Enhanced Continual Learning of Vision-Language Models with Model Fusion

Vision-Language Models (VLMs) represent a breakthrough in artificial intelligence by integrating visual and textual modalities to achieve impressive zero-shot capabilities. However, VLMs are susceptible to catastrophic forgetting when sequentially fine-tuned on multiple downstream tasks. Existing continual learning methods for VLMs often rely heavily on additional reference datasets, compromise zero-shot performance, or are limited to parameter-efficient fine-tuning scenarios. In this paper, we propose Continual Decoupling-Unifying (ConDU), a novel approach, by introducing model fusion into continual learning for VLMs. ConDU maintains a unified model along with task triggers and prototype sets, employing an iterative process of decoupling task-specific models for previous tasks and unifying them with the model for the newly learned task. Additionally, we introduce an inference strategy for zero-shot scenarios by aggregating predictions from multiple decoupled task-specific models. Extensive experiments across various settings show that ConDU achieves up to a 2\% improvement in average performance across all seen tasks compared to state-of-the-art baselines, while also enhancing zero-shot capabilities relative to the original VLM.

@article{gao2025_2503.10705,
  title={ Enhanced Continual Learning of Vision-Language Models with Model Fusion },
  author={ Haoyuan Gao and Zicong Zhang and Yuqi Wei and Linglan Zhao and Guilin Li and Yexin Li and Linghe Kong and Weiran Huang },
  journal={arXiv preprint arXiv:2503.10705},
  year={ 2025 }
}