AI Paper Atlas

Research map / Image generation / Editing and style transfer

Multi-turn Consistent Image Editing

Zijun Zhou, Yingying Deng, Xiangyu He, Weiming Dong, Fan Tang

ICCV 2025 · Image generation › Editing and style transfer · approach: Localized and attribute-specific editing · 23 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 multi-turn composite LQR target (vs single-LQR) (ablation)Ours · PIE-Bench fourth turnFID: 9.886 -> 5.553Improved
add multi-turn composite LQR target (vs single-LQR) (ablation)Ours · PIE-Bench fourth turnCLIP-T: 26.484 -> 26.831Mixed
add attention injection (vs w/o attn) (ablation)Ours · PIE-Bench fourth turnFID: 6.678 -> 5.553Improved
add attention injection (vs w/o attn) (ablation)Ours · PIE-Bench fourth turnCLIP-I: 0.889 -> 0.894Improved
use default rather than high attention weights (ablation)Ours · PIE-Bench fourth turnFID: 6.316 -> 5.553Improved
use default rather than high attention weights (ablation)Ours · PIE-Bench fourth turnCLIP-T: 26.878 -> 26.831No clear effect
use Ours instead of RF-InversionRF-Inversion · PIE-Bench fourth turnCLIP-T: 24.094 -> 26.831Improved
use Ours instead of FlowEditFlowEdit · PIE-Bench fourth turnFID: 14.547 -> 5.553Improved
use Ours instead of RF-SolverRF-Solver · PIE-Bench fourth turnFID: 11.581 -> 5.553Improved
use Ours (8 steps) instead of 15 stepsOurs · PIE-Bench fourth turnFID: 5.553 -> 5.396Improved
use Ours instead of FireFlow-vFireFlow-v · PIE-Bench fourth turnCLIP-T: 28.281 -> 26.831Hurt
apply multi-turn LQR in first editing roundOurs · PIE-Bench first turnFID: 2.554 vs 0.998 (FlowEdit)Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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