MoEpic splits each MoE expert into a cached top segment and a prefetched bottom segment, raising cache hit rates and hiding loading latency to accelerate offloaded MoE inference.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Accelerating Mixture-of-Expert Inference with Adaptive Expert Split Mechanism
MoEpic splits each MoE expert into a cached top segment and a prefetched bottom segment, raising cache hit rates and hiding loading latency to accelerate offloaded MoE inference.