原书出处
Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)
What this shows is that Wq and Wk define some matrix that modulates how we measure similarity (dot product) between ti and tj. A single self-attention layer therefore measures similarity in just one way.
原书定位详情
来源:a19c7933-4eda-4176-aebd-30fe7e1ad292
正文位置:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c026.xhtml
EPUB SHA-256:60a7eebc43f5d9a675d776fed3736fbdaa791e55877023f2f8cba016a9b384f8