ZeNOVA：流形约束的初始噪声优化方法实现高效生成模型对齐
Manifold-Constrained Initial Noise Optimization for Efficient Generative Model Alignment
AI 导读
arXiv 论文提出 ZeNOVA，一种无梯度的初始噪声对齐方法，通过退火软值引导、流形约束超球面 Langevin 动力学和 Metropolis-Hastings 跳跃，解决黑盒奖励场景下一阶梯度方法不稳定或不可用的问题。该方法面向一步或少步生成的蒸馏与流图模型，在图像和视频生成模型实验中表现优于所有评测的零阶基线，优化初始噪声更稳定地获得更高奖励。论文共 25 页、13 幅图。
正文 · 原文
Abstract:Recent advances in distillation and flow-map models have enabled deterministic one- or few-step generation for high-quality data, facilitating a new branch of reward alignment approaches that directly optimize the initial noise from a Gaussian distribution. However, most existing initial-noise optimization methods rely on first-order gradient information, which is either inapplicable or suffers from instability and inefficiency in black-box reward scenarios. Here, we introduce ZeNOVA, a stable and efficient initial noise alignment method in a gradient-free manner. Specifically, we address existing algorithms' major challenge in black-box scenarios through annealed soft-value guidance, manifold-constrained hyperspherical Langevin dynamics, and Metropolis-Hastings jumping. Extensive experiments on image and video generative models show that ZeNOVA outperforms all evaluated zeroth-order baselines by optimizing the initial noise toward higher rewards substantially more stably while exploiting the geometry of the Gaussian prior, demonstrating its practical applicability to various black-box reward alignment.
| Comments: | 25 pages, 13 figures |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2610.00365 [cs.LG] |
| (or arXiv:2610.00365v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.00365 arXiv-issued DOI via DataCite |
Submission history
From: Jong Chul Ye [view email]
[v1]
Wed, 30 Sep 2026 06:37:54 UTC (49,194 KB)
[v2]
Thu, 5 Nov 2026 05:23:11 UTC (49,194 KB)
来源：arXiv cs.AI · arxiv.org