In June 2017 Google Brain published the Transformer. Eight years on, every serious language model still descends from one short paper.
Home