CoLoGen: Progressive Learning of Concept`-`Localization Duality for Unified Image Generation
2026-02-25 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors address the challenge of creating one system that can handle many types of image generation tasks, which usually need different kinds of information. They introduce CoLoGen, a method that gradually teaches the system to understand both concepts and exact locations in images, then combines these skills for complex tasks. Their approach uses a special module, called PRW, to send information to expert parts and merge their results smoothly. Tests show CoLoGen works well for editing, controlled creation, and customized image generation.
conditional image generationdiffusion modelsconceptual representationlocalization cuescurriculum learningProgressive Representation Weavingimage editingcontrollable generation
Authors
YuXin Song, Yu Lu, Haoyuan Sun, Huanjin Yao, Fanglong Liu, Yifan Sun, Haocheng Feng, Hang Zhou, Jingdong Wang
Abstract
Unified conditional image generation remains difficult because different tasks depend on fundamentally different internal representations. Some require conceptual understanding for semantic synthesis, while others rely on localization cues for spatial precision. Forcing these heterogeneous tasks to share a single representation leads to concept`-`localization representational conflict. To address this issue, we propose CoLoGen, a unified diffusion framework that progressively learns and reconciles this concept`-`localization duality. CoLoGen uses a staged curriculum that first builds core conceptual and localization abilities, then adapts them to diverse visual conditions, and finally refines their synergy for complex instruction`-`driven tasks. Central to this process is the Progressive Representation Weaving (PRW) module, which dynamically routes features to specialized experts and stably integrates their outputs across stages. Experiments on editing, controllable generation, and customized generation show that CoLoGen achieves competitive or superior performance, offering a principled representational perspective for unified image generation.