"CTA: Hardware-Software Co-design for Compressed Token Attention Mechanism."

Haoran Wang et al. (2023)

Details and statistics

DOI: 10.1109/HPCA56546.2023.10070997

access: closed

type: Conference or Workshop Paper

metadata version: 2023-05-23

a service of  Schloss Dagstuhl - Leibniz Center for Informatics