Trajectory-Guided Diffusion for Foreground-Preserving Background Generation in Multi-Layer Documents

TL;DR

This paper introduces a diffusion-based method for generating multi-page document backgrounds that preserves foreground content and maintains stylistic consistency without additional constraints, by manipulating latent space trajectories.

Contribution

It presents a novel latent-space trajectory approach for diffusion models that ensures foreground preservation and style consistency across multiple pages without retraining.

Findings

Produces coherent multi-page backgrounds with preserved foregrounds.

Ensures stylistic consistency across pages using cached style directions.

Compatible with existing diffusion models and training-free.

Abstract

We present a diffusion-based framework for document-centric background generation that achieves foreground preservation and multi-page stylistic consistency through latent-space design rather than explicit constraints. Instead of suppressing diffusion updates or applying masking heuristics, our approach reinterprets diffusion as the evolution of stochastic trajectories through a structured latent space. By shaping the initial noise and its geometric alignment, background generation naturally avoids designated foreground regions, allowing readable content to remain intact without auxiliary mechanisms. To address the long-standing issue of stylistic drift across pages, we decouple style control from text conditioning and introduce cached style directions as persistent vectors in latent space. Once selected, these directions constrain diffusion trajectories to a shared stylistic subspace, ensuring consistent appearance across pages and editing iterations. This formulation eliminates the need for repeated prompt-based style specification and provides a more stable foundation for multi-page generation. Our framework admits a geometric and physical interpretation, where diffusion paths evolve on a latent manifold shaped by preferred directions, and foreground regions are rarely traversed as a consequence of trajectory initialization rather than explicit exclusion. The proposed method is training-free, compatible with existing diffusion backbones, and produces visually coherent, foreground-preserving results across complex documents. By reframing diffusion as trajectory design in latent space, we offer a principled approach to consistent and structured generative modeling.