AI Paper Atlas

Research map / Image generation / Editing and style transfer

Self-Prompting Diffusion Transformer for Open-Vocabulary Scene Text Edit via In-Context Learning

Hongxi Li, Tong Wang, WU CHENGJING, Tianbao Liu, Jiangtao Yao, Xiaochao Qu, Xinxiao Wu, Luoqi Liu, Ting Liu

ICML 2026 · Image generation › Editing and style transfer · approach: Instruction-based diffusion editing

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 style prompts and cooldown training on paired data (ablation)self-prompting MMDiT · AnyText-benchmark EnglishFID: 11.56 -> 7.62Improved
add style prompts and cooldown training on paired data (ablation)self-prompting MMDiT · AnyText-benchmark ChineseFID: 12.01 -> 7.95Improved
add style prompts and cooldown training on paired data (ablation)self-prompting MMDiT · AnyText-benchmark EnglishSentence accuracy: 0.8738 -> 0.8857Improved
add style prompts and cooldown training on paired data (ablation)self-prompting MMDiT · AnyText-benchmark EnglishLPIPS: 0.0608 -> 0.0365Improved
add glyph prompt (rendered target text) (ablation)MMDiT pretraining · AnyText-benchmark EnglishSeq. ACC: 0.4127 -> 0.8738Improved
add style prompt during self-supervised pretraining on top of glyph prompt (ablation)MMDiT pretraining · AnyText-benchmark EnglishSeq. ACC: 0.8738 -> 0.5412Hurt
cooldown with paired data when using glyph and style prompts (ablation)MMDiT · AnyText-benchmark EnglishSeq. ACC: 0.5440 -> 0.7982Improved
cooldown with unpaired data when using glyph prompt only (ablation)MMDiT · AnyText-benchmark EnglishSeq. ACC: 0.8738 -> 0.8788No clear effect
add style prompt without glyph prompt (ablation)MMDiT pretraining · AnyText-benchmark EnglishSeq. ACC: 0.4127 -> 0.5236Mixed
add more languages sequentially to training (ablation)multilingual MMDiT · MST-EditSeq. ACC: Arabic 0.6812 -> 0.7596Improved
use mask-guided formulation instead of instruction-based general editorsself-prompting MMDiT vs FireRed-Image-Edit-1.1 · AnyText-benchmark EnglishSeq. Acc: 0.7375 -> 0.8857Improved
text length above 16 charactersself-prompting MMDiT · text length 2 to 20 charactersvisual qualityHurt

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