AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

CoLoGen: Progressive Learning of Concept-Localization Duality for Unified Image Generation

Yuxin Song, Yu Lu, Haoyuan Sun, Huanjin Yao, Fanglong Liu, Yifan Sun, Haocheng Feng, Hang Zhou, Jingdong Wang

CVPR 2026 · Image generation › Text-to-image diffusion · approach: Personalized and region-controlled generation · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add concept representation (Rc) experts (ablation)CoLoGen · MagicBrush editingCLIP-T: 0.260 -> 0.302Improved
add localization representation (Rl) (ablation)CoLoGen · MagicBrush editingCLIP-I: 0.889 -> 0.922Improved
add localization representation (Rl) (ablation)CoLoGen · DreamBenchDINO: 0.683 -> 0.707Improved
add concept representation only (ablation)CoLoGen · MagicBrush editingCLIP-I: 0.889 -> 0.881Hurt
progressive staged training instead of multi-task co-training (ablation)CoLoGen · MagicBrush editingCLIP-T: 0.269 -> 0.301Improved
progressive staged training instead of multi-task co-training (ablation)CoLoGen · DreamBenchDINO: 0.679 -> 0.714Improved
multi-task co-training of Rc and Rl vs baseline (ablation)CoLoGen · DreamBenchDINO: 0.683 -> 0.679Hurt
add veteran gate routing supervision (ablation)CoLoGen · MagicBrush editing (instruction-editing stage)CLIP-IImproved
routing density rho 0.8 vs other values (ablation)CoLoGen · MagicBrush editingCLIP-IImproved
CoLoGen vs strongest controllable baselines on ADE20K segmentationCoLoGen · Seg-to-Image (ADE20K)mIoU: 44.23 -> 42.82Hurt
CoLoGen vs UniReal on editing l1CoLoGen · Emu Edit testl1: 0.099 -> 0.111Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 4 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy