Kushagra Pandey

PhD Student, University of California, Irvine


Hi, I am currently a second year PhD student in Computer Science at the University of California, Irvine advised by Prof. Stephan Mandt. I recently completed a summer internship supervised by Dr. Maja Rudolph at Bosch AI Research. My current research work is supported by the HPI Fellowship.

Previously, I was a graduate student in Computer Science at the Indian Institute of Technology, Kanpur advised by Prof. Piyush Rai where I also collaborated with Dr. Abhishek Kumar and Prof. Hamim Zafar.

My current research interests broadly lie in the study of Deep Generative Models and Unsupervised Representation Learning. I am currently primarily interested in the theory and applications of Score-Based (or Diffusion) Models and related families (like Flow Matching, Interpolants etc). Feel free to contact me about any collaboration opportunities or queries about my work in general.


Jan 27, 2024 Our new work on designing Efficient Integrators for Diffusion Generative Models has been accepted at ICLR’24. Huge shoutout to my advisors Stephan and Maja for their advising on this project.
Oct 27, 2023 Our work on “Towards Fast Stochastic Sampling in Diffusion Generative Models” has been accepted at the NeurIPS 2023 Workshop on Diffusion Models :sparkles:
Oct 12, 2023 Check out our new work on designing Efficient Integrators for Diffusion Generative Models. We devise efficient frameworks for fast deterministic/stochastic sampling from diffusion models.
Oct 07, 2023 Attended ICCV’23 in Paris. Check out our work on A Complete Recipe for Diffusion Generative Models which was accepted as an Oral Presentation at ICCV’23.