Browsing NTNU Open by Journals "2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)"
Now showing items 1-1 of 1
-
Paramixer: Parameterizing Mixing Links in Sparse Factors Works Better than Dot-Product Self-Attention
(Journal article, 2022)Self-Attention is a widely used building block in neural modeling to mix long-range data elements. Most self-attention neural networks employ pairwise dot-products to specify the attention coefficients. However, these ...