arXiv Computer Vision research summaries for May 23, 2024.

Today's Research Themes (AI-Generated):

• Unveiling inconsistencies in Large Vision-Language Models with novel benchmark tool ConBench.

• Efficient historical document categorization through semantic segmentation and deep learning embeddings.

• Assessing Vision-Large-Language-Models' vulnerability to typographic adversarial attacks in autonomous driving.

• Introducing SAM priors into multi-view remote sensing image segmentation for enhanced performance.

• Exploring Multi-Scale Vision Mamba for high-efficiency, long-range dependency learning in vision tasks.