Attention Modules

Multi-Head Attention

Introduced by Vaswani et al. in Attention Is All You Need

Multi-head Attention is a module for attention mechanisms which runs through an attention mechanism several times in parallel. The independent attention outputs are then concatenated and linearly transformed into the expected dimension. Intuitively, multiple attention heads allows for attending to parts of the sequence differently (e.g. longer-term dependencies versus shorter-term dependencies).

$$ \text{MultiHead}\left(\textbf{Q}, \textbf{K}, \textbf{V}\right) = \left[\text{head}_{1},\dots,\text{head}_{h}\right]\textbf{W}_{0}$$

$$\text{where} \text{ head}_{i} = \text{Attention} \left(\textbf{Q}\textbf{W}_{i}^{Q}, \textbf{K}\textbf{W}_{i}^{K}, \textbf{V}\textbf{W}_{i}^{V} \right) $$

Above $\textbf{W}$ are all learnable parameter matrices.

Note that scaled dot-product attention is most commonly used in this module, although in principle it can be swapped out for other types of attention mechanism.

Source: Lilian Weng

Source: Attention Is All You Need


Paper Code Results Date Stars


Task Papers Share
Language Modelling 67 8.86%
Retrieval 34 4.50%
Question Answering 30 3.97%
Large Language Model 27 3.57%
In-Context Learning 18 2.38%
Sentence 17 2.25%
Text Generation 15 1.98%
Semantic Segmentation 11 1.46%
Machine Translation 11 1.46%