arXiv：cs.LG（机器学习，全量分类）· Jinho Chang, Jong Chul Ye·· 14 小时前AI 评分31
ZeNOVA：面向生成模型对齐的流形约束初始噪声优化
Manifold-Constrained Initial Noise Optimization for Efficient Generative Model Alignment
AI 导读
ZeNOVA 是一种无需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现有算法不稳定、低效的问题。在图像和视频生成模型实验中，ZeNOVA 在更高奖励优化上稳定超越所有被评估的零阶基线，并利用高斯先验的几何结构，适用于各类黑盒奖励对齐。
正文
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:2610.00365v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.00365 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jong Chul Ye [view email]
[v1]
Wed, 30 Sep 2026 06:37:54 UTC (49,194 KB)
来源：arXiv：cs.LG（机器学习，全量分类） · arxiv.org