OBJECT RECORD / AIM-DOC-2017-001
Attention Is All You Need
The paper that introduced the Transformer architecture.
CURATORIAL READING
Why it is here
The authors proposed an encoder–decoder architecture built around attention, without recurrence or convolution. The object matters because it gives a compact technical account of the architecture rather than a retrospective origin story.
Evidence boundary
Start with Figure 1 and the model architecture section. Separate what the paper actually proposes from everything later products built on top of it.
11 / CHANGE ARCHITECTURE
Position in the guided path
From written rules to learned representations
The paper’s encoder–decoder uses attention without recurrence or convolution.
Back to the object room →PROVENANCE
Source trail
arXiv ·