The article discusses the Transformer family, a set of neural network architectures introduced in 2017 by Vaswani et al. Since its introduction, the Transformer has become a widely used and influential model in natural language processing tasks such as machine translation, text classification, and question answering. The author updates this post to include new models that have been developed since 2020. --- Why it matters: This matters to AI researchers because the Transformer family's impact on NLP tasks is significant, with many state-of-the-art results achieved using these architectures. Source: https://lilianweng.github.io/posts/2020-04-07-the-transformer-family/