AI Paper Atlas

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

ChimeraLoRA: Multi-Head LoRA-Guided Synthetic Datasets

Hoyoung Kim, Minwoo Jang, Jabin Koo, Sangdoo Yun, Jungseul Ok

CVPR 2026 · Image generation › Text-to-image diffusion · approach: Text-driven synthesis and semantic guidance · 0 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 multi-head LoRA (shared A, per-image B) to per-image LoRA (ablation)LoFT-style baseline · FGVCAircraft 4-shotaccuracy (%): 41.7 -> 43.9Improved
add semantic boosting (bounding-box-preserving crop) to per-image LoRA (ablation)LoFT-style baseline · FGVCAircraft 4-shotaccuracy (%): 41.7 -> 44.4Improved
combine multi-head LoRA and semantic boosting (ablation)ChimeraLoRA · FGVCAircraft 4-shotaccuracy (%): 41.7 -> 46.0Improved
add semantic boosting on top of multi-head LoRA (ablation)ChimeraLoRA · Flowers102 4-shotaccuracy (%): 93.1 -> 93.4Improved
add semantic boosting without multi-head LoRA (ablation)LoFT-style baseline · Flowers102 4-shotaccuracy (%): 91.3 -> 92.2Improved
share LoRA B instead of LoRA A across images (ablation)ChimeraLoRA · qualitative (headphone, motorbike, stapler, panda)image fidelityHurt
add 500 synthetic images per class with ChimeraLoRA to 4-shot real dataCLIP ViT-B/16 fine-tuning · 4-shot, 9 datasets (average)accuracy (%): 71.8 -> 74.6Improved
add 500 synthetic images per class with DataDreamCLIP ViT-B/16 fine-tuning · 4-shot, 9 datasets (average)accuracy (%): 71.8 -> 71.3Hurt
add synthetic images to tail classes with ChimeraLoRACLIP ViT-B/16 long-tail training · EuroSAT long-tail (tail classes)accuracy (%): 13.6 -> 51.3Improved
increase synthetic images per class (ablation)DataDream and LoFT · EuroSAT, FGVCAircraftaccuracy (%)Hurt
increase synthetic images per class (ablation)ChimeraLoRA · EuroSAT, FGVCAircraftaccuracy (%)Improved
sample Dirichlet mixture weights per image vs uniform weights 1/K (ablation)ChimeraLoRA · synthesiscoverageImproved

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