Architecture
Attention Is All You Need
Vaswani et al. 2017 paper that introduced the Transformer architecture.
Definition
This NeurIPS 2017 paper showed that pure self-attention, without recurrence or convolution, could match state-of-the-art machine translation. It is the foundational reference for nearly every modern LLM and was the catalyst for the LLM era that followed.
Common use cases
- Historical reference
- Architecture education