arXiv (All)open access
EStream: Fast and Memory-Efficient MoE Prefill through Expert Virtualization on Mobile NPUs
distributed, parallel, and cluster computing, hardware architecture, machine learning
This document is indexed with metadata only — full text is not available in the archive for this record.
Open the official source ↗
Related documents
Record · ID 974665
Retrieved via
Conceptio — every document is proof-bundled with source, license, and retrieval metadata.