About

"Attention is all you need" is the title of a 2017 machine learning paper, that is sometimes jokingly referred to in other contexts as a catchphrase "X is all you need".

Origin

The landmark paper was written by Google Brain researcher Ashish Vaswani and his co-authors and marks one of the most important steps forwards in recent Deep Learning history. It introduces the Transformer architecture that is the basis for various seminal technological advances in language processing, image classification [https://paperswithcode.com/paper/an-image-is-worth-16×16-words-transformers-1] and generative models [https://paperswithcode.com/paper/transgan-two-transformers-can-make-one-strong]. OpenAI's notorious GPT-3, for example, is a Transformer-based model architecture ("Generative Pre-trained Transformer 3").

Notable Examples