A construction principle for diffusion reference processes that gives simulation-free training and finite-time generation simultaneously, with conditional flow matching recovered as the zero-noise limit.
Figure 2 shows generated samples for T= 1,10,100, where the number of discretization steps (i.e
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Simulation-free and finite-time diffusion model
A construction principle for diffusion reference processes that gives simulation-free training and finite-time generation simultaneously, with conditional flow matching recovered as the zero-noise limit.