Vishal Purohit

Hi! I am a 4^th year Ph.D. student in the ECE program at Purdue University, where I work with Prof. Qiang Qiu. My research focuses on advancing generative models, including flow-based, diffusion, and autoregressive approaches, as well as solving inverse problems in computer vision. I aim to develop efficient algorithms for learning high-dimensional signals from sparse and quantized measurements with real-world applications.

During my Ph.D. I have interned at AMD Artificial Intelligence Group under Dr. Zicheng Liu.

Previously, I worked as a research intern at Visual Computing and Analytics Lab (VCAL), Indian Institute of Technology (BHU) for an year, where I contributed to projects in meta-learning and federated learning under the guidance of Prof. Sanjay Kumar Singh. Before pursuing academia, I was an engineer in the Medical Electronics division at Tata Elxsi for 2 years, where I developed software and hardware solutions for medical devices, gaining practical experience in designing systems for critical applications.

• Generative Models (2D and 3D) - Autoregressive and Diffusion
• Multimodal Large Models and multi-agent generative systems
• Video understanding and representation learning