# SOWing Information: Cultivating Contextual Coherence with MLLMs in Image Generation Source: [SOWing Information: Cultivating Contextual Coherence with MLLMs in Image Generation]() Collection index: [llms.txt]() Data updated: 2026-05-09T05:00:04.794Z Authors: Yuhan Pei, Ruoyu Wang, Yongqi Yang, Ye Zhu, Olga Russakovsky, Yu Wu Published: 2024-11-28T14:35:25.000Z ## Abstract Originating from the diffusion phenomenon in physics, which describes the random movement and collisions of particles, diffusion generative models simulate a random walk in the data space along the denoising trajectory. This allows information to diffuse across regions, yielding harmonious outcomes. However, the chaotic and disordered nature of information diffusion in diffusion models often results in undesired interference between image regions, causing degraded detail preservation and contextual inconsistency. In this work, we address these challenges by reframing disordered diffusion as a powerful tool for text-vision-to-image generation (TV2I) tasks, achieving pixel-level condition fidelity while maintaining visual and semantic coherence throughout the image. We first introduce Cyclic One-Way Diffusion (COW), which provides an efficient unidirectional diffusion framework for precise information transfer while minimizing disruptive interference. Building on COW, we further propose Selective One-Way Diffusion (SOW), which utilizes Multimodal Large Language Models (MLLMs) to clarify the semantic and spatial relationships within the image. Based on these insights, SOW combines attention mechanisms to dynamically regulate the direction and intensity of diffusion according to contextual relationships. Extensive experiments demonstrate the untapped potential of controlled information diffusion, offering a path to more adaptive and versatile generative models in a learning-free manner. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Text-to-Visual Attribute Editing on CelebA TV2I (val)](): Text-to-Visual Image Generation (Attribute Editing); CelebA TV2I (val); ID-Distance: 0.8068. - [Text-and-Visual-to-Image Generation (Style Transfer) on CelebA TV2I 1.0 (test val)](): Text-and-Visual-to-Image Generation (Style Transfer); CelebA TV2I 1.0 (test val); ID Distance: 0.8068. - [Text-to-Image with Visual condition on CelebA-TV2I (test)](): Text-to-Image with Visual condition; CelebA-TV2I (test); ID-Distance: 0.771. ## Original sources and implementation [Original paper]() [PDF]()