TarFlow sampling is accelerated up to 5.32x using Gauss-Seidel-Jacobi iteration after ranking blocks by a Convergence Ranking Metric and choosing initial values with an Initial Guessing Metric, keeping FID within about one percent.
Neural autoregressive flows
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Accelerate TarFlow Sampling with GS-Jacobi Iteration
TarFlow sampling is accelerated up to 5.32x using Gauss-Seidel-Jacobi iteration after ranking blocks by a Convergence Ranking Metric and choosing initial values with an Initial Guessing Metric, keeping FID within about one percent.