"Mixture-of-Depths: Dynamically allocating compute in transformer-based ..."

David Raposo et al. (2024)

Details and statistics

DOI: 10.48550/ARXIV.2404.02258

access: open

type: Informal or Other Publication

metadata version: 2024-05-13

a service of  Schloss Dagstuhl - Leibniz Center for Informatics