← All works
Research paper

Attention Is All You Need

At a glance

The Transformer uses attention to translate text while allowing faster training.

In one paragraph

The Transformer uses attention to connect relevant words without processing a sentence one word at a time. This allowed more training to happen in parallel. The paper demonstrated better translation results with less training time than leading earlier approaches.

Authors 8

  1. Ashish Vaswani
  2. Noam Shazeer
  3. Niki Parmar
  4. Jakob Uszkoreit
  5. Llion Jones
  6. Aidan Gomez
  7. Lukasz Kaiser
  8. Illia Polosukhin

Organizations

Credited for this work at publication.