SOURCE-LINKED INTELLIGENCE

ToPO: Token-Conditioned Preference Routing for Attention-Based Latent Diffusion Models

arXiv · AI, language, vision and robotics · article · Sep 3, 2026 · UTC

Pairwise preference labels rank complete images, yet Diffusion-DPO applies their effect over many spatial and denoising-time coordinates. For attention-based, noise-prediction latent diffusion, ToPO (Token-Oriented Preference Optimization) constructs a per-minibatch, detached, separable spatial-temporal route from branchwise squared-residual contrast in a frozen reference denoiser. Preferred-branch cross-attention uses content tokens to modulate the spatial factor, and an auxiliary pixel-midpoint ordering term is added without local labels or a learned reward model. In matched three-seed retra

Read original source ↗ Open in workspace

recordType

paper

region

Global