Journal club;
         machine learning;
         scientific computing.

Deschutes Hall, University of Oregon, Eugene, Oregon
Exact location floating; 2pm; Fridays.

Current paper (2026.02.13)

• DiffusionSat: A Generative Foundation Model for Satellite Imagery
• Presenter - Michael Dushkoff

• Flow Matching for Generative Modeling
• Flow Matching with Gaussian Process Priors for Probabilistic Time Series Forecasting
• InverseBench: Benchmarking Plug-and-Play Diffusion Priors for Inverse Problems in Physical Sciences
• Rethinking FID: Towards a Better Evaluation Metric for Image Generation
• Distribution estimation via Flow Matching with Lipschitz guarantees
• On the minimax optimality of Flow Matching through the connection to kernel density estimation
• Inductive Moment Matching
• Improving Diffusion Inverse Problem Solving with Decoupled Noise Annealing
• CoCoGen: Physically-Consistent and Conditioned Score-based Generative Models for Forward and Inverse Problems
• Scalable Adaptive Computation for Iterative Generation
• Simplifying, Stabilizing and Scaling Continuous-Time Consistency Models
• Force Matching with Relativistic Constraints: A Physics-Inspired Approach to Stable and Efficient Generative Modeling
• Generative Modeling of Random Fields from Limited Data via Constrained Latent Flow Matching
• STARFlow: Scaling Latent Normalizing Flows for High-resolution Image Synthesis
• Boosting Latent Diffusion with Flow Matching
• Improving and generalizing flow-based generative models with minibatch optimal transport