Login / Signup

Effectively Compress KV Heads for LLM.

Hao YuZelan YangShen LiYong LiJianxin Wu
Published in: CoRR (2024)
Keyphrases
  • real time
  • high quality
  • transmission line
  • databases
  • learning algorithm
  • three dimensional
  • multiscale
  • natural language
  • management system
  • electron beam