AI Paper Atlas

Research map / Image generation / Editing and style transfer

Anchor Token Matching: Implicit Structure Locking for Training-free AR Image Editing

Taihang Hu, Linxuan Li, Kai Wang, Yaxing Wang, Jian Yang, Ming-Ming Cheng

ICCV 2025 · Image generation › Editing and style transfer · approach: Instruction-based diffusion editing · 5 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
replace naive prompt modification with anchor token matchingLlamaGen editing · PIE-Bench subset (5 editing types)Structure Distance: 113.95 -> 31.79Improved
replace attention-map injection (PnP-AR) with anchor token matchingLlamaGen editing · PIE-Bench subset (5 editing types)Structure Distance: 103.94 -> 31.79Improved
replace naive prompt modification with anchor token matchingLlamaGen editing · PIE-Bench subset (5 editing types)PSNR: 12.14 -> 19.75Improved
anchor token matching vs naive modification on whole-image CLIPLlamaGen editing · PIE-Bench subset (5 editing types)CLIP similarity (whole): 24.71 -> 24.19Hurt
use dynamic window instead of fixed |W|=50 (ablation)Anchor Token Matching · PIE-Bench subset (5 editing types)Structure Distance: 60.83 -> 31.79Improved
use dynamic window instead of fixed |W|=150 (ablation)Anchor Token Matching · PIE-Bench subset (5 editing types)CLIP similarity: 22.18 -> 24.19Improved
use dynamic window instead of fixed |W|=150 (ablation)Anchor Token Matching · PIE-Bench subset (5 editing types)Structure Distance: 30.39 -> 31.79Hurt
use dynamic window instead of fixed windows (ablation)Anchor Token Matching · PIE-Bench subset (5 editing types)Structure Distance / CLIP ratio: 1.37 -> 1.31Improved
increase fixed window size from 50 to 150 (ablation)Anchor Token Matching · PIE-Bench subset (5 editing types)Structure Distance: 60.83 -> 30.39Improved
increase fixed window size from 50 to 150 (ablation)Anchor Token Matching · PIE-Bench subset (5 editing types)CLIP similarity: 24.79 -> 22.18Hurt
lower threshold tau in adaptive constraint relaxation (ablation)Anchor Token Matching · qualitative editsediting strengthMixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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