Showing stat.MLShow all
3 papers · 1 filter
stat.ML2026
Implicit geometric regularization in flow matching via density weighted Stein operators
Shinto Eguchi
Flow Matching (FM) has emerged as a powerful paradigm for continuous normalizing flows, yet standard FM implicitly performs an unweighted regression over the entire ambient s…
stat.ML2026
Robust inference using density-powered Stein operators
Shinto Eguchi
We introduce a density-power weighted variant of the Stein operator, called the -Stein operator, for robust inference with unnormalized probability models. The operator is motiv…
stat.ML2026
Statistical Inference via Generative Models: Flow Matching and Causal Inference
Shinto Eguchi
Generative AI has achieved remarkable empirical success, but from the perspective of statistics it often remains opaque: its predictions may be accurate, yet the underlying mechani…