WePE encodes 2D patch positions in Vision Transformers via Weierstrass elliptic functions on the complex plane to exploit double periodicity and derive relative positions algebraically.
Bridging the gap between vision transformers and convolutional neural net- works on small datasets,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Weierstrass Positional Encoding for Vision Transformers
WePE encodes 2D patch positions in Vision Transformers via Weierstrass elliptic functions on the complex plane to exploit double periodicity and derive relative positions algebraically.