By designing the prior as the low-frequency projection of data images, flow matching achieves OT-optimal identity couplings without explicit OT computation, reducing trajectory curvature over 2x and improving few-step quality.
Alignflow: Improving flow-based generative models with semi-discrete optimal transport
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Optimal Transport Flow Matching by Design
By designing the prior as the low-frequency projection of data images, flow matching achieves OT-optimal identity couplings without explicit OT computation, reducing trajectory curvature over 2x and improving few-step quality.