Norm-Based Strategy for KV Cache Compression.

Alessio Devoto Yu Zhao Simone Scardapane Pasquale Minervini

Published in: CoRR (2024)

Keyphrases

video sequences
compression scheme
image compression
objective function
query processing
compression algorithm
data compression
back end
neural network
database systems
pattern matching
main memory
compression ratio
prefetching
hit rate
garbage collection