Cascade Prompt Learning for Vision-Language Model Adaptation [ECCV 2024]

Cascade Prompt Learning for Vision-Language Model Adaptation
Ge Wu*, Xin Zhang*, Zheng Li, Zhaowei Chen, Jiajun Liang, Jian Yang Xiang Li *Joint first authors

---

News

(March 2025) The Chinese version of the paper has been released in this link. Thanks for Liu Tao's efforts

(September 2024) Training codes of CasPL+PromptSRC will be released.

(July 2024) Our work is accepted to ECCV 2024!

---

Cascade Prompt Learning

Highlights

1. We propose a novel cascade prompt learning framework consisting of both boosting and adapting prompt phases. To our best knowledge, CasPL is the first to introduce cascaded diverse prompts with multiple phases for VLMs, which is a brand new learning paradigm for fine-tuning VLMs.
2. We demonstrate that the boosting prompts can distill domain-general knowledge from the senior teacher over massive unlabeled domain images, leading to superior recognition performance and efficient inference.
3. As a plug-and-play framework, CasPL can be seamlessly integrated into any existing prompt learning approaches, with negligible parameters (boosting prompt tokens, <0.1%) and ignorable additional inference cost introduced.

Framework

Figure 1. An overview of CasPL framework. (a) We utilize a set of boosting prompts to enable the student CLIP model to extract general domain knowledge from the teacher CLIP model, leveraging an extensive amount of unlabeled domain data. (b) The boosting prompt can be seamlessly incorporated into existing related work as a plug-in. Here, we exemplify this integration with PromptSRC, where frozen boosting prompts are cascaded with learnable adapting prompts without altering any loss function.

Results

Base-to-Novel: Results reported below show accuracy for base and novel classes for across 11 recognition datasets averaged over 3 seeds.

Few-shot: Results reported below show accuracy across 11 recognition datasets averaged over 3 seeds.

---

Running

Installation

For installation and other package requirements, please follow the instructions detailed in INSTALL.md.

Data Preparation

Please follow the instructions at DATASETS.md to prepare all datasets.

Training

(0) Get boosting prompt.

python train_script/train_get_boosting_prompt.py

Add data path and save path for get the weight of boosting_prompt.

dataset_list = [['fgvc_aircraft', "data_path/fgvc-aircraft-2013b/"]]
SAVE_PATH="save_path/caspl_teacher"

Or you can download the weight related to the boosting prompt on this link, and save it in save_path.

(1) Base-to-Novel Experiments.

Add data_path and save_path. The weight of base-to-novel can be found in this link.

dataset_list = [['fgvc_aircraft', "data_path/fgvc-aircraft-2013b/"]]
SAVE_PATH="save_path/caspl_promptsrc_base_to_novel"
SAVE_PATH_TEACHER="save_path/caspl_teacher"

Run the training script below.

python train_script/caspl_promptsrc/train_base_to_new.py

(2) Few-shots Experiments.

Add data path, save path, and use the script trained below. The weight of few-shots can be find in this link.

python train_script/caspl_promptsrc/train_few_shot.py

(3) Cross-dataset Experiments.

Add data path, save path, and use the script trained below. The weight of cross-dataset can be find in this link.

python train_script/caspl_promptsrc/train_cross_dataset.py

---

Citation

If you find our work, this repository, or pretrained models useful, please consider giving a star and citation.

@inproceedings{wu2025cascade,
  title={Cascade prompt learning for vision-language model adaptation},
  author={Wu, Ge and Zhang, Xin and Li, Zheng and Chen, Zhaowei and Liang, Jiajun and Yang, Jian and Li, Xiang},
  booktitle={European Conference on Computer Vision},
  pages={304--321},
  year={2025},
  organization={Springer}
}

Contact

If you have any questions, please create an issue on this repository or contact at gewu.nku@gmail.com.

Acknowledgements

Our code is based on PromptSRC, along with Co-CoOp and CoOp repository. We thank the authors for releasing their code. If you use our model and code, please consider citing these works as well.