Oracle Noise: Faster Semantic Spherical Alignment for Interpretable Latent Optimization

TL;DR

Oracle Noise introduces a Riemannian hypersphere-based zero-shot framework that accelerates semantic alignment in text-to-image diffusion models, improving quality and efficiency without external parsers.

Contribution

It redefines noise initialization as a spherical, semantic-driven optimization, preserving Gaussian priors and enabling faster, artifact-free alignment in generative models.

Findings

Achieves state-of-the-art performance on human preference metrics.

Significantly accelerates semantic alignment within 2 seconds.

Eliminates Euclidean norm inflation, reducing visual artifacts.

Abstract

Text-to-image diffusion models have achieved remarkable generative capabilities, yet accurately aligning complex textual prompts with synthesized layouts remains an ongoing challenge. In these models, the initial Gaussian noise acts as a critical structural seed dictating the macroscopic layout. Recent online optimization and search methods attempt to refine this noise to enhance text-image alignment. However, relying on unconstrained Euclidean gradient ascent mathematically inflates the latent norm and destroys the standard Gaussian prior, causing severe visual artifacts like color over-saturation. Furthermore, these methods suffer from inefficient semantic routing and easily fall into the ``reward hacking'' trap of external proxy models. To address these intertwined bottlenecks, we propose Oracle Noise, a zero-shot framework reframing noise initialization as semantic-driven optimization strictly confined to a Riemannian hypersphere. Instead of relying on complex external parsers, we directly identify the most impactful structural words in the prompt to efficiently route optimization energy. By updating the noise strictly along a spherical path, we mathematically preserve the original Gaussian distribution. This geometric constraint eliminates norm inflation and unlocks aggressive step sizes for rapid convergence. Extensive experiments demonstrate that Oracle Noise significantly accelerates semantic alignment and achieves superior aesthetics without black-box models. It completely mitigates Euclidean-induced degradation, establishing state-of-the-art performance across human preference metrics (e.g., HPSv2, ImageReward), semantic alignment (CLIP Score), and sample diversity, all within a strict 2-second optimization budget.