arXiv：cs.LG· Woobin Park, Jaeyeon Kim, Yilun Du, Sitan Chen, Bohyung Han·· 2 小时前
Priming Models：一步生成真的需要想那么复杂吗？
Are We Overthinking One-Step Generation?
AI 导读
Priming Models 是一种两阶段一步生成器训练框架，先在固定噪声级别下做像素空间去噪重建，再进行特征空间分布后训练。该重建阶段即可完全消除分布训练的难度，在 ImageNet 256×256 上以约十分之一的训练算力实现一步生成器 SOTA 的 FID 1.19。代码已开源。
正文
Abstract:Diffusion and flow matching are now dominant frameworks for generative modeling in continuous domains, but they typically require multiple network evaluations at inference. This has motivated extensive research on one-step generative modeling primarily built on diffusion and flow, though existing approaches often rely on complex training machinery. In this work, we introduce Priming Models, a simple two-stage framework for training one-step generators. Motivated by the difficulty of distributional training from scratch, Priming Models are first trained on a reconstruction objective, e.g., pixel-space denoising at a fixed noise level, before undergoing feature-space distributional post-training. Our central scientific finding is that this simple reconstruction stage is already enough to entirely alleviate the difficulty of distributional training. Empirically, across class-conditional image generation, text-to-image generation, and world modeling, Priming Models significantly advance the Pareto frontier of training compute vs. generation quality. On ImageNet at $256 \times 256$ resolution, we achieve a state-of-the-art FID of $1.19$ among one-step generators using approximately one-tenth the training compute of prior methods with comparable FID. We open-source our codebase at this https URL.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.20331 [cs.LG] |
| (or arXiv:2610.20331v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.20331 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jaeyeon Kim [view email]
[v1]
Wed, 14 Oct 2026 15:14:43 UTC (47,744 KB)
来源：arXiv：cs.LG · arxiv.org