In EdgeVLMOpt (EVO): Optimizing Vision-Language Models for Resource-Constrained Edge Devices, we aim to develop efficient and scalable techniques to enable the deployment of advanced vision-language models (VLMs) on edge hardware. While VLMs have demonstrated strong capabilities in multimodal reasoning and understanding, their high computational and memory demands pose significant challenges for real-time, on-device applications.

This project focuses on exploring model optimization strategies, including compression, quantization, and efficient architecture design, to reduce the resource footprint of VLMs while preserving their performance. By combining advances in deep learning, transformer architectures, and hardware-aware optimization, the project seeks to enable low-latency, privacy-preserving, and reliable multimodal intelligence directly on edge devices.

Researchers and students will investigate novel approaches to improve inference efficiency and adaptability, preparing them for cutting-edge applications in areas such as augmented reality, robotics, and intelligent edge systems. A practical example of this project includes, but is not limited to, deploying compact vision-language models for real-time multimodal understanding on embedded platforms.

Feel free to visit my website here and contact me for more information.