"Loss Masking Is Not Needed in Decoder-only Transformer for Discrete-token ..."

Qian Chen et al. (2023)

Details and statistics

DOI: 10.48550/ARXIV.2311.04534

access: open

type: Informal or Other Publication

metadata version: 2024-07-20

a service of  Schloss Dagstuhl - Leibniz Center for Informatics