Model techniques map
Techniquesmodel architecturetoken mixersparse attention

specific method · filed under model architecture

Native Sparse Attention

A sparse-attention method described as having compression, selection, sliding-window paths, and a learned gate.

Also called NSA.

source
1
model
1
labs adopt it
0
strongest
not used

How sources treat it

One count per evidence span, weakest treatment to strongest.

not used 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

NSA originally has three parallel paths (compress + select + sliding window) plus a learned gate. M3 keeps only selection.

not usedmodel architecturein MiniMax-M3MiniMax

Filed alongside

Other methods under model architecture :: token mixer :: sparse attention.