Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Pre-trained Model Guided Fine-Tuning for Zero-Shot Adversarial Robustness

About

Large-scale pre-trained vision-language models like CLIP have demonstrated impressive performance across various tasks, and exhibit remarkable zero-shot generalization capability, while they are also vulnerable to imperceptible adversarial examples. Existing works typically employ adversarial training (fine-tuning) as a defense method against adversarial examples. However, direct application to the CLIP model may result in overfitting, compromising the model's capacity for generalization. In this paper, we propose Pre-trained Model Guided Adversarial Fine-Tuning (PMG-AFT) method, which leverages supervision from the original pre-trained model by carefully designing an auxiliary branch, to enhance the model's zero-shot adversarial robustness. Specifically, PMG-AFT minimizes the distance between the features of adversarial examples in the target model and those in the pre-trained model, aiming to preserve the generalization features already captured by the pre-trained model. Extensive Experiments on 15 zero-shot datasets demonstrate that PMG-AFT significantly outperforms the state-of-the-art method, improving the top-1 robust accuracy by an average of 4.99%. Furthermore, our approach consistently improves clean accuracy by an average of 8.72%. Our code is available at https://github.com/serendipity1122/Pre-trained-Model-Guided-Fine-Tuning-for-Zero-Shot-Adversarial-Robustness.

Sibo Wang, Jie Zhang, Zheng Yuan, Shiguang Shan• 2024

Related benchmarks

TaskDatasetResultRank
Image ClassificationImageNet A
Top-1 Acc12.3
654
Image ClassificationImageNet V2
Top-1 Acc54.8
611
Image ClassificationSUN397
Accuracy55.31
441
Image ClassificationFGVCAircraft
Accuracy15.09
261
Image ClassificationImageNet-R--
217
Image ClassificationOxfordPets
Accuracy84.11
160
Image ClassificationCIFAR10
Top-1 Accuracy83.24
112
Image ClassificationCIFAR100
Accuracy43.94
102
Image ClassificationImageNet-S
Top-1 Acc38.4
92
Image ClassificationStanfordCars
Robust Accuracy29
91
Showing 10 of 173 rows
...

Other info

Follow for update