AI Paper Atlas

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

StyleKeeper: Prevent Content Leakage using Negative Visual Query Guidance

Jaeseok Jeong, Junho Kim, Gayoung Lee, Yunjey Choi, Youngjung Uh

ICCV 2025 · Image generation › Text-to-image diffusion · approach: Text-driven synthesis and semantic guidance · 1 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 CFG with swapping self-attention (ablation)StyleKeeper · T2I visual style promptingtext alignment / image quality (qualitative)Improved
add negative visual query guidance (NVQG) (ablation)StyleKeeper · T2I visual style promptingcontent leakage (qualitative)Improved
swap self-attention only in upblocks instead of all blocks (ablation)StyleKeeper · T2I visual style promptingcontent leakage (qualitative)Improved
swap self-attention in bottleneck blocks (ablation)StyleKeeper · T2I visual style promptingcontent leakage (qualitative)Hurt
start swapping in late upblock layers (24th layer onward) instead of early (ablation)StyleKeeper · T2I visual style promptingstyle / leakage trade-off (qualitative)Improved
use stochastic encoding instead of DDIM inversion for real references (ablation)StyleKeeper · real image referenceartifacts / style similarity (qualitative)Improved
add color calibration (ablation)StyleKeeper · real image referencestyle similarity and text alignment (qualitative)Improved
StyleKeeper vs StyleAlignedstyle similarity · Gram-matrix style similaritystyle similarity (Gram): 0.759 -> 0.791Improved
StyleKeeper vs IP-Adapterstyle similarity · Gram-matrix style similaritystyle similarity (Gram): 0.768 -> 0.791Improved
StyleKeeper vs IP-Adapteruser preference · User studyvote share: 18.47% -> 58.15%Improved
StyleKeeper vs StyleAligneduser preference · User studyvote share: 13.15% -> 58.15%Improved
apply stochastic encoding and color calibration to other self-attention variants (StyleAligned, MasaCtrl)existing methods · real image referenceartifacts / style (qualitative)Improved

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