AI Paper Atlas

Research map / Image generation / Editing and style transfer

CogniEdit: Dense Gradient Flow Optimization for Fine-Grained Image Editing

Yan Li, Lin Liu, Xiaopeng Zhang, Wei Xue, Wenhan Luo, Yike Guo, Qi Tian

CVPR 2026 · Image generation › Editing and style transfer · approach: Instruction-based diffusion editing · 2 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 Dense GRPO optimization (ablation)CogniEdit (Qwen-Image-Edit) · KRIS-Bench Knowledge ReasoningAvg score: 46.11 -> 59.29Improved
add Dynamic Token Focus Relocation (ablation)CogniEdit (Qwen-Image-Edit) · KRIS-Bench Knowledge ReasoningAvg score: 54.66 -> 59.29Improved
add both components to Qwen-Image-Edit-r1 (ablation)Qwen-Image-Edit-r1 · KRIS-Bench Knowledge ReasoningAvg score: 55.48 -> 59.29Improved
add Dynamic Token Focus Relocation without Dense GRPO (ablation)Qwen-Image-Edit-r1 · KRIS-Bench Knowledge ReasoningAvg score: 55.48 -> 46.11Hurt
add Dense GRPO without Dynamic Token Focus Relocation (ablation)Qwen-Image-Edit-r1 · KRIS-Bench Knowledge ReasoningAvg score: 55.48 -> 54.66No clear effect
add Dense GRPO and dynamic focus to Qwen-Image-Edit-r1 (ablation)Qwen-Image-Edit-r1 · KRIS-Bench Knowledge ReasoningIF: 35.03 -> 40.14Improved
add VLM instruction reasoning (r1) to Qwen-Image-EditQwen-Image-Edit · KRIS-Bench Social ScienceAvg score: 66.40 -> 77.99Improved
train CogniEdit on top of Qwen-Image-Edit-r1Qwen-Image-Edit-r1 · KRIS-Bench Social ScienceAvg score: 77.99 -> 77.64No clear effect
train CogniEdit on top of Qwen-Image-Edit-r1Qwen-Image-Edit-r1 · KRIS-Bench Natural ScienceAvg score: 64.78 -> 67.42Improved
Dense GRPO versus standard GRPO (ablation)CogniEdit · CLIP score vs FLOPsCLIP scoreImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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