FINCH: Prompt-guided Key-Value Cache Compression for Large Language Models

FINCH: Prompt-guided Key-Value Cache Compression for Large Language Models Giulio Corallo author Paolo Papotti author 2024 text journal article Transactions of the Association for Computational Linguistics continuing MIT Press Cambridge, MA periodical academic journal corallo-papotti-2024-finch 10.1162/tacl_a_00716 https://aclanthology.org/2024.tacl-1.83/ 2024 12 1517 1532