A buffer-free MoE dispatch and combine method on Ascend hardware with pooled HBM cuts intermediate relay overhead via direct expert window access.
Deepspeed-moe: Advancing mixture-of-experts inference and training to power next-generation ai scale
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Relay Buffer Independent Communication over Pooled HBM for Efficient MoE Inference on Ascend
A buffer-free MoE dispatch and combine method on Ascend hardware with pooled HBM cuts intermediate relay overhead via direct expert window access.