AI Paper Atlas

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

Contrastive Test-Time Composition of Multiple LoRA Models for Image Generation

Tuna Han Salih Meral, Enis Simsar, Federico Tombari, Pinar Yanardag

ICCV 2025 · Image generation › Text-to-image diffusion · approach: Multi-subject personalization · 13 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
CLoRA vs strongest baseline Mix-of-ShowStable Diffusion v1.5 multi-LoRA · 131 LoRAs, 200 two-subject promptsAvg DINO similarity: 0.526 -> 0.554Improved
CLoRA vs strongest baseline Mix-of-ShowStable Diffusion v1.5 multi-LoRA · 131 LoRAs, 200 two-subject promptsAvg CLIP-I: 0.712 -> 0.725Improved
CLoRA vs Composite on text alignmentStable Diffusion v1.5 multi-LoRA · 131 LoRAs, 200 two-subject promptsCLIP-T: 0.833 -> 0.862Improved
CLoRA vs ZipLoRA in user studyStable Diffusion v1.5 multi-LoRA · 131 LoRAs, 200 two-subject promptsUser study faithfulness (1-5): 2.81 -> 3.32Improved
CLoRA vs Mix-of-Show on segmented instancesStable Diffusion v1.5 multi-LoRA · SEEM instance-segmented compositionsAvg CLIP similarity (%): 78.7% -> 87.1%Improved
CLoRA vs Mix-of-Show and Orthogonal Adaptation with 3-5 LoRAsStable Diffusion v1.5 multi-LoRA · 3-5 LoRA compositionsAvg DINO similarity: 0.351 -> 0.347No clear effect
increase LoRAs from 2 to 8CLoRA · scalabilityruntime / VRAM: 24s -> 96s; 25GB -> 80GBHurt
add Latent Update (contrastive objective) (ablation)CLoRA · qualitativecomposition correctnessImproved
add Latent Masking (ablation)CLoRA · qualitativesubject identityImproved

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