Model Tells You Where to Merge: Adaptive KV Cache Merging for LLMs on Long-Context Tasks.

Zheng Wang Boxiao Jin Zhongzhi Yu Minjia Zhang

Published in: CoRR (2024)

Keyphrases

computational model
formal model
cost function
prior knowledge
probability distribution
management system
petri net
mathematical model
experimental data
hybrid model
database
neural network
machine learning
em algorithm
conceptual model
conceptual framework