Dithering Defense: Adversarial Robustness of Vision Foundation Models via Multi-Level Floyd-Steinberg Dithering
Researchers have developed a new method called multi-level Floyd-Steinberg error-diffusion dithering to enhance the adversarial robustness of vision foundation models. This technique acts as an input transformation that disrupts adversarial attacks while maintaining the semantic content of the images. Tested across various tasks and model families, the dithering method, particularly with intermediate quantization and post-processing blur, demonstrated superior or comparable performance to existing baselines with less degradation on clean inputs. AI
IMPACT Introduces a lightweight, model-agnostic defense against adversarial attacks for vision foundation models.