Tecoa2 Clip

🚀 Open Clip

Open Clip is a library that provides a TeCoA CLIP ViT - L/14 model for zero - shot image classification, which is fine - tuned on ImageNet using supervised adversarial methods.

🚀 Quick Start

This library offers a TeCoA CLIP ViT - L/14 model, initialized from OpenAI CLIP and fine - tuned on ImageNet with supervised adversarial methods using infinity - norm and a radius of 2/255.

✨ Features

• Based on the TeCoA (Mao et al. (2023)) framework.
• Supervised adversarial fine - tuning on ImageNet for enhanced performance.

💻 Usage Examples

Basic Usage

model, _, image_processor = open_clip.create_model_and_transforms('hf - hub:chs20/tecoa2 - clip')

📚 Documentation

[Paper] [GitHub]

📄 License

This project is licensed under the MIT license.

📚 Citation

If you find this model useful, please consider citing our paper:

@article{schlarmann2024robustclip,
    title={Robust CLIP: Unsupervised Adversarial Fine - Tuning of Vision Embeddings for Robust Large Vision - Language Models}, 
    author={Christian Schlarmann and Naman Deep Singh and Francesco Croce and Matthias Hein},
    year={2024},
    journal={ICML}
}