AI Paper Atlas

Research map / Architectures and efficiency / CNN design

A Single Architecture for Representing Invariance Under Any Space Group

Cindy Zhang, Elif Ertekin, Peter Orbanz, Ryan P Adams

ICLR 2026 · Architectures and efficiency › CNN design · approach: Group symmetry equivariant networks · 1 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 symmetry-adapted Fourier positional encoding to the Transformer (ablation)Transformer · Materials Project total energyMAE: 0.220 -> 0.197Improved
add symmetry-adapted Fourier positional encoding to the Transformer (ablation)Transformer · Materials Project band gapMAE: 0.440 -> 0.306Improved
add symmetry-adapted Fourier positional encoding to the Transformer (ablation)Transformer · Materials Project bulk modulusMAE: 0.142 -> 0.082Improved
add symmetry-adapted Fourier positional encoding to the Transformer (ablation)Transformer · Materials Project shear modulusMAE: 0.226 -> 0.158Improved
use CFT instead of ALIGNNALIGNN · Materials Project total energyMAE: 0.203 -> 0.197Improved
use CFT instead of ALIGNNALIGNN · Materials Project band gapMAE: 0.235 -> 0.306Hurt
use CFT instead of MatformerMatformer · Materials Project total energyMAE: 0.371 -> 0.197Improved
use CFT instead of MatformerMatformer · Materials Project bulk modulusMAE: 0.074 -> 0.082No clear effect
use CFT instead of CGCNNCGCNN · Materials Project band gapMAE: 0.343 -> 0.306Improved
use CFT instead of ALIGNN for training speedALIGNN · Materials Project efficiencytraining time per epoch: 592 sec -> 91 secImproved
use CFT instead of ALIGNN for inference speedALIGNN · Materials Project efficiencyinference time: 451 sec -> 60 secImproved
use CFT instead of Matformer for training speedMatformer · Materials Project efficiencytraining time per epoch: 266 sec -> 91 secImproved

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