Model techniques map
Techniquesinference & servinginference kernel

implementation detail · filed under inference & serving

Fused QSA kernel

A kernel that jointly computes sparse-attention outputs and KL loss without materializing intermediate results.

Also called Fused sparse-attention and KL-loss kernel.

source
1
model
1
lab adopt it
1
strongest
used

How sources treat it

One count per evidence span, weakest treatment to strongest.

used 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

we implement a fused QSA kernel that jointly computes sparse attention outputs and the KL loss without materializing intermediate results

usedunclearin Qwen3.8-Flash-NextQwen

Filed alongside

Other methods under inference & serving :: inference kernel.