AI Paper Atlas

Research map / Image generation / Editing and style transfer

StableI2I: Spotting Unintended Changes in Image-to-Image Transition

Jiayang Li, Shuo Cao, Xiaohui Li, Zhizhen Zhang, Kaiwen Zhu, Yule Duan, Yu Qiao, Jian Zhang, Yihao Liu

ICML 2026 · Image generation › Editing and style transfer · approach: Instruction-based diffusion editing · 0 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
apply RL to the base model (ablation)Qwen3-VL-8B-Instruct · Binary & Type QA averageaccuracy: 33.17 -> 64.95Improved
add SFT before RL (ablation)StableI2I RL · Binary & Type QA averageaccuracy: 64.95 -> 80.47Improved
add multiple-choice QA in the RL stage only (ablation)StableI2I SFT+RL · Binary & Type QA averageaccuracy: 80.47 -> 80.73No clear effect
add multiple-choice QA in the SFT stage as well as RL (ablation)StableI2I SFT+RL · Binary & Type QA averageaccuracy: 80.73 -> 81.90Improved
add multiple-choice QA in both stages (ablation)StableI2I SFT+RL · Binary & Type QA averageaccuracy: 80.47 -> 81.90Improved
add multiple-choice QA in both stages (ablation)StableI2I SFT+RL · Binary & Type QA low-levelaccuracy: 94.10 -> 95.00Improved
add RL on augmented data (Dev.2 vs Dev.1) (ablation)StableI2I-Dev.1 · Binary & Type QA semanticaccuracy: 59.00 -> 67.70Improved
add RL on augmented data (Dev.2 vs Dev.1) (ablation)StableI2I-Dev.1 · Binary & Type QA structureaccuracy: 83.70 -> 80.30Hurt
re-train with augmented plus original data (final vs Dev.1) (ablation)StableI2I-Dev.1 · StableI2I-Bench totalaccuracy: 85.58 -> 87.35Improved
re-train with augmented plus original data (final vs Dev.2) (ablation)StableI2I-Dev.2 · StableI2I-Bench totalaccuracy: 86.35 -> 87.35Improved
add multi-stage training and data augmentation (ablation)Qwen3-VL-8B-Instruct · StableI2I-Bench totalaccuracy: 37.53 -> 87.35Improved
add multi-stage training on multiple-choice QA (ablation)Qwen3-VL-8B-Instruct · Multiple-choice QA averageaccuracy: 41.90 -> 91.70Improved

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