A matrix ISA extension that stores matrix tiles in ordinary vector registers and lets software choose tile geometry at runtime reports 1.35x average speedup over Intel AMX in simulation.
Efficient direct convo- lution using long simd instructions,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.AR 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
A Flexible Instruction Set Architecture for Efficient GEMMs
A matrix ISA extension that stores matrix tiles in ordinary vector registers and lets software choose tile geometry at runtime reports 1.35x average speedup over Intel AMX in simulation.