AI Paper Atlas

Research map / Image generation / Editing and style transfer

Training-Free Multi-Concept Image Editing

Niki Maria Foteinopoulou, Ignas Budvytis, Stephan Liwicki

ECCV 2026 · Image generation › Editing and style transfer · approach: Instruction-based diffusion editing · 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
combine ordered timesteps, regularisation and negative prompts over DDS (ablation)CDS · InstructPix2PixCLIPScore: 0.305 -> 0.308Improved
add only strict ordered timesteps to DDS (ablation)CDS · InstructPix2PixLPIPS: 0.264 -> 0.461Hurt
add only regularisation to DDS (ablation)CDS · InstructPix2PixCLIPScore: 0.305 -> 0.291Hurt
add only regularisation to DDS (ablation)CDS · InstructPix2PixLPIPS: 0.264 -> 0.081Improved
use CDS objective instead of PDSCDS · InstructPix2PixCLIPScore: 0.298 -> 0.308Improved
use dynamic concept weighting instead of MergeCDS · ComposLoRA RealityLPIPS: 0.610 -> 0.474Improved
use dynamic concept weighting instead of SwitchCDS · ComposLoRA RealityLPIPS: 0.473 -> 0.474No clear effect
use CDS instead of Composite on CLIPScoreCDS · ComposLoRA RealityCLIPScore: 0.365 -> 0.359Hurt
use CDS instead of SwitchCDS · ComposLoRA AnimeLPIPS: 0.345 -> 0.314Improved
use smaller patch size (2x2 vs 16x16) in concept weighting (ablation)CDS · ComposLoRA subsetCLIPScore: 0.332 -> 0.351Improved
lower temperature tau from 1 to 0.002 (ablation)CDS · ComposLoRA subsetCLIPScore: 0.344 -> 0.351Improved
increase regularisation strength eta from 0.01 to 10 (ablation)CDS · InstructPix2PixLPIPS: 0.407 -> 0.089 (CLIP 0.316 -> 0.299)Mixed

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