Publications

Robust Vision Transformers: Defending Against Adversarial Attacks in Critical Applications

A. Chen, S. Williams, E. Rodriguez, J. Liu

IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2025)

We present a comprehensive framework for enhancing the adversarial robustness of Vision Transformers in safety-critical applications. Our approach combines adversarial training with architectural modifications to achieve state-of-the-art robustness while maintaining competitive accuracy.

Interpretable Medical Image Analysis: A Multi-Modal Approach to Trustworthy Diagnosis

J. Liu, S. Williams, P. Patel, A. Chen

Medical Image Computing and Computer Assisted Intervention (MICCAI 2025)

This work introduces a novel interpretable framework for medical image analysis that combines multiple imaging modalities while providing clinicians with transparent decision explanations. Our method achieves superior diagnostic accuracy while maintaining full interpretability.