SOURCE-LINKED INTELLIGENCE

Gaussian Flow-Matching Schedules: Implications for Sampling and Training

arXiv · AI, language, vision and robotics · article · Sep 22, 2026 · UTC

Flow-matching schedules affect both sampling dynamics and the variance of the regression target. For centered commuting Gaussians, we show that a direction-dependent schedule decomposes into two independent design choices: a variance path, which fully determines the intermediate laws and probability flow, and a factorization, which leaves this flow unchanged while controlling irreducible regression variance. On the sampling side, we analyze finite-step Euler accuracy and derive a necessary drift bound for exact N -step sampling, connecting the geodesic and the logarithmic path. On the training

Read original source ↗ Open in workspace

recordType

paper

region

Global