Concept record · edition 1.0.0
Transformers
What the term means here
A sequence architecture built on attention rather than recurrence or convolution.
Why it matters
Architectural properties set what can be scaled and what the inference cost profile looks like.
Related claim records · 1
Sources used here · 1
primary paperContent verified
Attention Is All You Need
Vaswani et al. · 2017
The architecture paper underlying the transformer models this atlas is about.
Source record →