AI Paper Atlas

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

ConceptSplit: Decoupled Multi-Concept Personalization of Diffusion Models via Token-wise Adaptation and Attention Disentanglement

Habin Lim, Yeongseob Won, Juwon Seo, Gyeong-Moon Park

ICCV 2025 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance · 3 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
train only value projection instead of key and value (ablation)ToVA · 2 objects, no backgroundDINO image alignment: 0.474 -> 0.566Improved
train only value projection instead of key only (ablation)ToVA · 1 object, no backgroundDINO image alignment: 0.175 -> 0.809Improved
add LODA stage 1 latent optimization (ablation)ConceptSplit, 2 objects · 2 objects, no backgroundGenEval: 0.550 -> 0.825Improved
add LODA stage 1 latent optimization (ablation)ConceptSplit, 3 objects · 3 objects, no backgroundGenEval: 0.012 -> 0.026No clear effect
add attention fixing guidance (stage 2) (ablation)ConceptSplit · 2 objects, no backgroundGenEval: 0.825 -> 0.902Improved
add attention fixing guidance (stage 2) (ablation)ConceptSplit · 3 objects, no backgroundGenEval: 0.026 -> 0.344Improved
ConceptSplit versus best baselines on 2 objects with backgroundmulti-concept personalization · 2 objects, with backgroundGenEval: 0.288 -> 0.648Improved
ConceptSplit on single object with no backgroundsingle-concept personalization · 1 object, no backgroundGenEval: 0.954 -> 0.942No clear effect
merging-free token-wise value adapters versus EDLoRA adapter mergingtraining cost, three concepts · training timetraining time: ~84m -> ~9mImproved
modify key projection of cross-attention (ablation)personalization methods · attention entropy change, dog tokenentropy change: -0.003600 -> -0.000266Hurt

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