Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, & Illia Polosukhin (2017)
arXiv.
DOI: https://doi.org/10.48550/arxiv.1706.03762
Abstract. Introduces the Transformer, an encoder-decoder architecture based entirely on self-attention with no recurrence or convolution. The Transformer achieved state-of-the-art machine translation and became the foundational architecture for modern language, vision, and multimodal AI.
Tags: transformer attention foundational
Cited in: