MANO replaces quadratic self-attention with multiscale windowed attention over progressively downsampled grids, keeping complexity linear and reporting competitive accuracy on vision and PDE benchmarks.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Linear Attention with Global Context: A Multipole Attention Mechanism for Vision and Physics
MANO replaces quadratic self-attention with multiscale windowed attention over progressively downsampled grids, keeping complexity linear and reporting competitive accuracy on vision and PDE benchmarks.