A landing algorithm lets diffusion models sample from nonconvex constrained sets using underdamped dynamics without repeated projections or Newton solves.
N−1X k=0 lnp θ(xk|xk+1, xk+2) # (Training loss-ULLA-P) ≤E q(x0:N)EρN(pN |xN)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Efficient Diffusion Models under Nonconvex Equality and Inequality constraints via Landing
A landing algorithm lets diffusion models sample from nonconvex constrained sets using underdamped dynamics without repeated projections or Newton solves.