Efficient Boltzmann Sampling
Efficient sampling from Boltzmann distributions using consistency models and importance sampling.
I combined consistency models with importance sampling to sample Boltzmann distributions efficiently.
The method reduced function evaluations from 100 to 6–25 while preserving effective sample size on synthetic and equivariant n-body systems.
Related publication: Efficient and Unbiased Sampling of Boltzmann Distributions via Consistency Models.