GitHub

DART: Detect Anything in Real Time

Training-free framework that converts SAM3 into a real-time multi-class open-vocabulary detector. Achieves 55.8 AP on COCO val2017 (80 classes) at 15.8 FPS (4 classes, 1008px) on a single RTX 4080.
IEEE

Automated Code Generation: A CNN-Based Alternative to Large Language Models Using Multi-Head Mixed Convolution and Depthwise Separable Neural Networks

Abstract: As intelligent behavior becomes a core expectation of software, automatic code generation has gained significant attention. While Large Language Models (LLMs) dominate recent research, they ...
IEEE

Malware Detection Using Transfer Learning with ResNet-50: An Image-Based Approach

Malware, or malicious software, is software that is designed to harm users, organizations, and telecommunication and computer systems. More specifically, malware can block internet connections, ...
AutoGuide

How Nissan is Using AI to Develop New Cars Faster

Artificial Intelligence (AI) is everywhere these days. With computers smarter than ever and the associated intelligence only getting better by the second, automakers are employing said tech to help ...