Safe Pretraining of Deep Language Models in a Synthetic Pseudo-Language › Research Explorer

DOI

https://doi.org/10.1134/S1064562423701636
Final published version

T. E. Gorbacheva
I. Y. Bondarenko

Original language	English
Pages (from-to)	S494-S502
Number of pages	9
Journal	Doklady Mathematics
Volume	108
Issue number	Suppl 2
DOIs	https://doi.org/10.1134/S1064562423701636
Publication status	Published - Dec 2023

Research areas

AI safety, automatic text generation, deep learning methods, language models, pretraining, synthetic data, transformers

OECD FOS+WOS

1.01 MATHEMATICS

ID: 59887909