A new technique from ETH Zurich accelerates language models by 300x; it is a modification of Google’s BERT transformer model November 27, 2023 // by Finnovate This content is for members only. Sign up for access to the latest trends and innovations in fintech. View subscription plans. Login