Citation:
+## Citation:
If you find this project useful please cite:
**EvoSeed**
+ # **EvoSeed**
Source for the article: [Breaking Free: How to Hack Safety Guardrails in Black-Box Diffusion Models!](https://arxiv.org/abs/2402.04699)
-> 
(New!) Added Tutorial to generate adversarial images for ResNet-50 using Stable Diffusion.
+> (New!) Added Tutorial to generate adversarial images for ResNet-50 using Stable Diffusion.
-## 
Key Contributions:
+## Key Contributions:
- We propose a black-box algorithmic framework based on an Evolutionary Strategy titled EvoSeed to generate natural adversarial samples in an unrestricted setting.
- Our results show that adversarial samples created using EvoSeed are photo-realistic and do not change the human perception of the generated image; however, can be misclassified by various robust and non-robust classifiers.
@@ -27,7 +27,7 @@
Figure: Adversarial images created with EvoSeed are prime examples of how to deceive a range of classifiers tailored for various tasks.
Note that, the generated natural adversarial images differ from non-adversarial ones, suggesting the adversarial images' unrestricted nature.
-## 
Tutorial:
+## Tutorial:
Tutorial for creating adversarial images for ResNet-50 using Stable Diffusion can be found in the [notebook](./code/Tutorial.ipynb)
Citation:
+## Citation:
If you find this project useful please cite: