Ge²mS-T introduces multi-dimensional grouped computation with ExpG-IF neurons and GW-SSA attention to jointly optimize memory, accuracy, and energy in spiking vision transformers.
Advancing spiking neural networks towards deep residual learning.IEEE Transactions on Neural Net- works and Learning Systems, 36(2):2353–2367
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.NE 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Ge$^\text{2}$mS-T: Multi-Dimensional Grouping for Ultra-High Energy Efficiency in Spiking Transformer
Ge²mS-T introduces multi-dimensional grouped computation with ExpG-IF neurons and GW-SSA attention to jointly optimize memory, accuracy, and energy in spiking vision transformers.