Conditional Generative
Conditional generative modeling focuses on creating new data instances conditioned on specific inputs, aiming to learn complex conditional distributions and generate high-fidelity samples. Current research emphasizes diverse model architectures, including diffusion models, energy-based models, and normalizing flows, often applied to tasks like image generation, video prediction, and scientific simulation. These advancements are significantly impacting various fields, enabling data augmentation for improved classification, realistic simulations in physics and medicine, and the development of more powerful AI tools for diverse applications.
Papers
October 30, 2023
October 27, 2023
October 20, 2023
October 19, 2023
October 18, 2023
October 7, 2023
October 5, 2023
October 4, 2023
September 30, 2023
August 3, 2023
July 31, 2023
July 28, 2023
June 25, 2023
June 22, 2023
June 20, 2023
June 18, 2023
May 27, 2023