Manifold-Constrained Initial Noise Optimization for Efficient Generative Model Alignment

Excerpt

arXiv:2610.00365v1 Announce Type: new 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