arxiv
PublishedMay 25, 2026 at 4:00 AM
Variance Reduction for Expectations with Diffusion Teachers
Publisher summary· verbatim
arXiv:2605.21489v2 Announce Type: replace-cross Abstract: Pretrained diffusion models serve as frozen teachers feeding downstream pipelines such as text-to-3D, single-step distillation, and data attribution. The teacher gradients these pipelines consume are Monte Carlo (MC) expectations over noise l
Stay posted· Newsletter
A 5-min weekly brief — top movers, price watch, story of the week.
Discussion
No replies yet. Be first.
The Bubble Brief
WEEKLYRead AI insights every Tuesday — top movers, new releases, story of the week.
Originally published on arxiv ↗