AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

LoCA: Spatially-Aware Low-Rank Convolutional Adaptation of Vision Foundation Models

Sojung An, Junha Lee, Sujeong You, Nam Ik Cho, Donghyun Kim

ECCV 2026 · Architectures and efficiency › Pruning and efficient tuning · approach: LoRA variants and theory · 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
apply flattened convolutional LoRA instead of linear-only LoRA (ablation)LoRA on ConvNeXt-L · GTAV to Cityscapes/BDD/Mapillary DGSSaverage mIoU: 65.29 -> 64.95Hurt
add low-rank channel mixing to LoRA baseline (ablation)LoCA on ConvNeXt-L · GTAV to Cityscapes/BDD/Mapillary DGSSaverage mIoU: 65.29 -> 66.16Improved
add SVD-based spatial basis refinement (ablation)LoCA on ConvNeXt-L · GTAV to Cityscapes/BDD/Mapillary DGSSaverage mIoU: 66.16 -> 67.03Improved
add hierarchical rank scheduling (ablation)LoCA on ConvNeXt-L · GTAV to Cityscapes/BDD/Mapillary DGSSaverage mIoU: 67.03 -> 67.46Improved
use covariance-SVD initialization instead of zero initialization (ablation)LoCA · VTAB-1k averageaverage accuracy: 75.7 -> 75.9Improved
use covariance-SVD initialization instead of flattened SVD (ablation)LoCA · VTAB-1k averageaverage accuracy: 73.0 -> 75.9Improved
use covariance-SVD initialization instead of zero initialization (ablation)LoCA · DGSS averageaverage mIoU: 65.6 -> 66.5Improved
initialize channel mixing with SoMA principal singular components instead of zero (ablation)LoCA on ConvNeXt-L · GTAV to Cityscapes/BDD/Mapillary DGSSaverage mIoU: 66.49 -> 67.46Improved
initialize channel mixing with SoMA principal singular components instead of zero (ablation)LoCA on ConvNeXt-B · GTAV to Cityscapes/BDD/Mapillary DGSSaverage mIoU: 63.76 -> 63.32Hurt
use LoCA instead of full fine-tuningMambaVision-B · GTAV to Cityscapes/BDD/Mapillary DGSSaverage mIoU: 32.52 -> 44.20Improved
use LoCA instead of full fine-tuningMobileMamba · ImageNet Top-1 accuracytop-1 accuracy: 70.6% -> 78.9%Improved
use LoCA r16 instead of r4 (ablation)LoCA on ConvNeXt-B · VTAB-1k (19 tasks)average accuracy: 75.98 -> 75.93No clear effect

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