AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Concept-Aware LoRA for Domain-Aligned Segmentation Dataset Generation

Minho Park, Sunghyun Park, Jungsoo Lee, Hyojin Park, Kyuwoong Hwang, Fatih Porikli, Jaegul Choo, Sungha Choi

CVPR 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Domain-generalized segmentation

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
select layers by concept awareness (top 2%) instead of LoRA on all projection layers (ablation)CA-LoRA · Cityscapes 0.3% in-domainmIoU: 42.97 -> 44.13Improved
fine-tune a random 2% subset of layers instead of concept-aware selection (ablation)CA-LoRA · Cityscapes 0.3% in-domainmIoU: 44.13 -> 43.24Hurt
add diffusion-gradient normalization to concept awareness (ablation)CA-LoRA · Cityscapes 0.3% in-domainmIoU: 43.01 -> 44.13Improved
fine-tune only Q projections (hand-crafted) (ablation)DatasetDM with LoRA · Cityscapes 0.3% in-domainmIoU: 42.82 -> 40.50Hurt
fine-tune only K projections (hand-crafted) (ablation)DatasetDM with LoRA · Cityscapes 0.3% in-domainmIoU: 42.82 -> 43.50Improved
measure concept awareness at timestep 81 instead of 1, 201 or 481 (ablation)CA-LoRA · Cityscapes 0.3% in-domainmIoU: 41.68 (t=201), 40.80 (t=481), 40.49 (t=1) -> 44.13 (t=81)Improved
increase the proportion of fine-tuned layers from 2% to 10% or 100% (ablation)CA-LoRA · Cityscapes 0.3% in-domainmIoU: 44.13 -> 43.36 (10%), 42.97 (100%)Hurt
use projection-wise grouping instead of block-wise, layer-wise or integrated projection-wise (ablation)CA-LoRA · Cityscapes 0.3% in-domainmIoU: 43.67 (block-wise) -> 44.13Improved
use Viewpoint CA-LoRA rather than Style CA-LoRA for domain generalization (ablation)CA-LoRA · adverse-weather text adherenceCLIP-ScoreImproved
fine-tune the T2I model with CA-LoRA before training the label generatorDatasetDM · Cityscapes in-domainmIoU: 42.82 -> 44.13 (0.3%)Improved
standard LoRA and AdaLoRA fine-tuning of the T2I modelDatasetDM · Cityscapes in-domainmIoU: AdaLoRA 3%: +1.86; 100%: -0.78Mixed
InstructPix2Pix generated dataMask2Former · Cityscapes in-domainmIoU: 10%: -2.81; 3%: +1.36Mixed

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