Abstract: Programming based approaches to reasoning tasks have substantially expanded the types of questions models can answer about visual scenes. Yet on benchmark visual reasoning data, when models ...

Like all AI models based on the Transformer architecture, the large language models (LLMs) that underpin today’s coding ...

Abstract: Recent studies show that the state-of-the-art deep neural networks (DNNs) are vulnerable to adversarial examples, resulting from small-magnitude perturbations added to the input. Given that ...

Participate in Minnesota-based prosopagnosia research. In the Prosopagnosia and Visual Cognition Lab at Bethel University, Dr. Sherryse Corrow and her students study high-level visual processes—like ...