AI Paper Atlas

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

Think in Strokes, Not Pixels: Process-Driven Image Generation via Interleaved Reasoning

Lei Zhang, Junjiao Tian, Zhipeng Fan, Kunpeng Li, Jialiang Wang, Weifeng Chen, Markos Georgopoulos, Felix Juefei-Xu, Julian McAuley, Manling Li, Zecheng He

ECCV 2026 · Image generation › Text-to-image diffusion · approach: T2I evaluation and controllable adapters · 7 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 instruction augmentation (refine, remove, swap) to multi-turn data (ablation)BAGEL-7B process-driven · GenEvalPosition: 0.58 -> 0.67Improved
add instruction augmentation to multi-turn data (ablation)BAGEL-7B process-driven · GenEvalColor Attributes: 0.50 -> 0.62Improved
add self-critique after instruction augmentation (ablation)BAGEL-7B process-driven · GenEvalPosition: 0.67 -> 0.72Improved
add self-critique without augmentation (ablation)BAGEL-7B process-driven · GenEvalPosition: 0.58 -> 0.61Improved
add scene-graph based critiques (ablation)BAGEL-7B process-driven · GenEvalPosition: 0.67 -> 0.70Improved
use self-sampled critiques instead of scene-graph critiques (ablation)BAGEL-7B process-driven · GenEvalColors / Position / Color Attr.: 0.83 / 0.70 / 0.67 -> 0.87 / 0.72 / 0.69Improved
add instruction-intermediate conflict supervision (ablation)BAGEL-7B process-driven · GenEvalPosition: 0.66 -> 0.71Improved
add image-instruction alignment supervision (ablation)BAGEL-7B process-driven · GenEvalCounting: 0.61 -> 0.73Improved
add image-instruction alignment supervision (ablation)BAGEL-7B process-driven · GenEvalPosition: 0.66 -> 0.69Mixed
combine both reasoning subsets (ablation)BAGEL-7B process-driven · GenEvalCounting / Position: 0.61 -> 0.75; 0.66 -> 0.72Improved
add process-driven interleaved reasoning fine-tuning to BAGEL-7BBAGEL-7B · GenEvalOverall: 0.77 -> 0.83Improved
add process-driven interleaved reasoning fine-tuning to BAGELBAGEL · WISEOverall: 0.70 -> 0.76Improved

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