MAS-Attention pipelines tiled MatMul and softmax workloads onto separate MAC and vector units of edge NPUs, achieving up to 2.75x simulated and 1.76x measured speedups over FLAT while keeping outputs exact.
https://docs.nvidia.com/deeplearning/tensorrt/archives/tensorrt-803/best-practices/index.html
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
MAS-Attention: Memory-Aware Stream Processing for Attention Acceleration on Resource-Constrained Edge Devices
MAS-Attention pipelines tiled MatMul and softmax workloads onto separate MAC and vector units of edge NPUs, achieving up to 2.75x simulated and 1.76x measured speedups over FLAT while keeping outputs exact.