Login / Signup

NeuPIMs: NPU-PIM Heterogeneous Acceleration for Batched LLM Inferencing.

Guseul HeoSangyeop LeeJaehong ChoHyunmin ChoiSanghyeon LeeHyungkyu HamGwangsun KimDivya MahajanJongse Park
Published in: CoRR (2024)
Keyphrases
  • knowledge representation
  • objective function
  • heterogeneous data
  • heterogeneous networks
  • data sets
  • three dimensional