Constructs G-equivariant ViTs for arbitrary discrete G ≤ O(2), proves H ≤ G implies G-models embed into H-models and single-head equivariant attention realizes all ordinary G-equivariant maps, introduces D6 hexagonal model, and reports preliminary accuracy gains on PatternNet in low-data regimes.
Stronger vits with octic equivariance
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2representative citing papers
Learning rotation invariance in descriptors matches the performance of matcher-level invariance but allows earlier invariance, faster matchers, and no loss in upright performance when trained at scale.
citing papers explorer
-
A Unified Framework for Vision Transformers Equivariant to Discrete Subgroups of $\mathrm{O}(2)$
Constructs G-equivariant ViTs for arbitrary discrete G ≤ O(2), proves H ≤ G implies G-models embed into H-models and single-head equivariant attention realizes all ordinary G-equivariant maps, introduces D6 hexagonal model, and reports preliminary accuracy gains on PatternNet in low-data regimes.
-
Who Handles Orientation? Investigating Invariance in Feature Matching
Learning rotation invariance in descriptors matches the performance of matcher-level invariance but allows earlier invariance, faster matchers, and no loss in upright performance when trained at scale.