Search Results for author:
Found 3 papers, 2 papers with code
GRAM: Global Reasoning for Multi-Page VQA
The increasing use of transformer-based large language models brings forward the challenge of processing long sequences.
Classifier Robustness Enhancement Via Test-Time Transformation
We show that the proposed method achieves state-of-the-art results and validate our claim through extensive experiments on a variety of defense methods, classifier architectures, and datasets.
Threat Model-Agnostic Adversarial Defense using Diffusion Models
Deep Neural Networks (DNNs) are highly sensitive to imperceptible malicious perturbations, known as adversarial attacks.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
