AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Spatial Priors via Space Filling Curves for Small and Limited Data Vision Transformers

Leyla Naz Candogan, Arshia Afzal, Pol Puigdemont, Volkan Cevher

ICML 2026 · Architectures and efficiency › Attention and transformers · approach: Hybrid and interpretable ViT designs · 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 Violin SFC decay mask at fine-tuningDeiT-T · VTAB-1K full fine-tuningavg top-1 accuracy (%): 65.52 -> 68.33Improved
add Violin SFC decay mask at fine-tuningDeiT-III-S · VTAB-1K Structuredaccuracy (%): 52.92 -> 61.61Improved
add Violin mask at fine-tuningDINO-S · VTAB-1K full fine-tuningavg top-1 accuracy (%): 71.21 -> 71.84Improved
add Violin mask at fine-tuningDeiT-III-H · VTAB-1K full fine-tuningavg top-1 accuracy (%): 66.91 -> 68.50Improved
add Violin mask on top of LoRALoRA · VTAB-1K PEFTavg accuracy (%): 71.04 -> 72.55Improved
add Violin mask on top of DoRADoRA · VTAB-1K PEFTavg accuracy (%): 70.75 -> 71.90Improved
add Violin to pixel-level ViTDeiT-T · Pixel-level CIFAR-100avg accuracy (%): 60.8 -> 68.0Improved
add Violin during ImageNet pretrainingDeiT-T · ImageNet-1Ktop-1 accuracy: +0.8%Improved
add Violin to Swin-T Mask R-CNNSwin-T · COCO detectionmAP: +0.4Improved
use four-curve Violin instead of a single Peano curve (ablation)Violin · VTAB-1K Structuredavg accuracy (%): 61.63 -> 61.89Improved
use multiplicative averaged mask instead of additive (ablation)Violin · VTAB-1K Structuredavg accuracy (%): 61.34 -> 61.89Improved
use SFC masks instead of Manhattan distance mask (ablation)Violin · VTAB-1K Structuredavg accuracy (%): 58.37 -> 61.89Improved

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