The field of computer vision is moving towards the development of more efficient models that can be deployed on resource-constrained devices. Recent research has focused on improving the performance of vision transformers, which have shown impressive results in various computer vision tasks. However, their computational demands often make them unsuitable for deployment on edge devices. To address this, researchers have proposed novel architectures and techniques such as knowledge distillation, pruning, and quantization to reduce the computational complexity of these models. These advancements have led to significant improvements in the efficiency and accuracy of computer vision models, enabling their deployment in real-world applications such as aerial object detection, crop monitoring, and autonomous vehicles. Notable papers in this area include CoSwin, which proposes a novel feature-fusion architecture for small-scale vision tasks, and BATR-FST, which introduces a bi-level adaptive token refinement approach for few-shot learning. Additionally, papers such as A Novel Compression Framework for YOLOv8 and Cott-ADNet have demonstrated the effectiveness of compression techniques and lightweight architectures for real-time object detection and image classification tasks.