Verlustfreie Kompression von Large Language Models

Near-Lossless Compression of Large Language Models

Dan Alistarh (ORCID: 0000-0003-3650-940X)

Wissenschaftsdisziplinen

Informatik (100%)

Keywords

  • Large Language Models,
  • Compression,
  • Efficiency,
  • Inference
Forschungsstätte(n)
  • Institute of Science and Technology - Austria (ISTA) - 100%
Nationale Projektbeteiligte
  • Eldar Kurtic, Institute of Science and Technology - Austria (ISTA) , nationale:r Kooperationspartner:in
Internationale Projektbeteiligte