| title | Posterior Matching Distillation |
|---|---|
| authors | Hidir Yesiltepe, Julius Berner, Arash Vahdat, Pinar Yanardag |
| date | 2026-11-19 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2611.26629 |
| affiliation | NVIDIA |
| upvotes | 31 (Hugging Face Daily Papers, 2026-11-21) |
| links | abs pdf hf code |
| cite | |
| related | Cosmos 3: Omnimodal World Models for Physical AI LocateAnything: Fast and High-Quality Vision-Language Grounding with Parallel Box Decoding LongLive-2.0: An NVFP4 Parallel Infrastructure for Long Video Generation SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning SoL-Pi: Recursively Scaling Auto-Research Loops for Efficient Agent Harness |
| abstract | Diffusion distillation accelerates image and video generation by compressing a pretrained teacher's iterative sampling process into a few network evaluations. Distribution-based methods achieve this by matching the student's output distribution to the teacher's. This matching is often driven by reverse-KL objectives, whose mode-seeking bias favors dominant teacher modes at the expense of diversity across samples. We introduce Posterior Matching Distillation (PMD), a new member of the distribution-based distillation family that combines mode-covering posterior sampling with any-step generation. At zero population loss, PMD recovers the teacher's full posterior of clean images at each trained noise level. Its stochastic student supports any-step sampling budgets within the trained timestep grid without further training. To learn these posteriors, PMD matches exponentially reweighted averages of student samples conditioned on a fixed noisy state to corresponding teacher posterior means. Under Gaussian corruption, exponentially tilted posterior mean is a single teacher evaluation at a shifted input. We prove that, matching tilted means over an open neighborhood uniquely identifies the full posterior. PMD thus supervises the entire conditional distribution through a family of mean constraints, without any auxiliary student-score network. Experiments on SDXL and FLUX.2 [klein] 4B demonstrate that PMD achieves the best FID among 4-step methods on both models, while maintaining competitive or leading performance across prompt-alignment and perceptual-quality metrics. |
bibtex
@misc{yesiltepe2026posterior,
title={Posterior Matching Distillation},
author={Hidir Yesiltepe and Julius Berner and Arash Vahdat and Pinar Yanardag},
year={2026},
eprint={2611.26629},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2611.26629}
}