AI Paper Atlas

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

ReFlex: Text-Guided Editing of Real Images in Rectified Flow via Mid-Step Feature Extraction and Attention Adaptation

Jimyeong Kim, Jungwon Park, Yeji Song, Nojun Kwak, Wonjong Rhee

ICCV 2025 · Image generation › Text-to-image diffusion · approach: Personalized and region-controlled generation · 14 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
use mid-step feature extraction instead of extracting at every early step of the inverted latent (ablation)ReFlex · PIE-Bench / Wild-TI2I-Realsource IoUImproved
add I2T-CA adaptation (ablation)ReFlex · PIE-Bench / Wild-TI2I-Realtext alignmentImproved
add I2I-SA adaptation (ablation)ReFlex · PIE-Bench / Wild-TI2I-Realtext alignmentImproved
extract features at t'=14 instead of fully inverted latent t'=28 (ablation)ReFlex · qualitativesource structure: t'=28 -> t'=14Improved
extract features at t'=14 instead of t'=4 (ablation)ReFlex · qualitativeimage quality: t'=4 -> t'=14Improved
increase k in I2I-SA adaptation (ablation)ReFlex · qualitativeeditability vs structureMixed
increase alpha from 1 in I2T-CA adaptation (ablation)ReFlex · qualitativetext alignmentMixed
apply latent blending for T steps instead of 0.7T (ablation)ReFlex (k=20) · PIE-BenchBackground PSNR: 24.21 -> 28.05Improved
apply latent blending for T steps instead of 0.7T (ablation)ReFlex (k=20) · PIE-BenchCLIP similarity (whole): 26.51 -> 26.30Hurt
increase k from 20 to 40 (ablation)ReFlex (m=0.7T) · PIE-BenchCLIP similarity (edited region): 23.17 -> 23.37Improved
increase k from 20 to 40 (ablation)ReFlex (m=0.7T) · PIE-BenchBackground PSNR: 24.21 -> 24.03Hurt
use ReFlex on FLUX instead of RF-EditFLUX editing · PIE-BenchCLIP similarity (whole): 25.41 -> 26.51Improved

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