New Study Proposes Optimal Interpolant Schedules for Generative Modeling
A new paper introduces 'elliptic sampling,' deriving optimal interpolant schedules for stochastic interpolant-based generative models, with proven optimality for Gaussians and extensions beyond.
Read the full story at arXiv stat.ML ↗
Timeline · 1 report
- 2026-11-16 05:00 · arXiv stat.ML
Elliptic Sampling: Optimal Interpolant Schedules for Gaussians and Beyond