arXiv - CSCL: "Alternatives to the Scaled Dot Product for Attent…" - CleverLibre Social

arXiv - CSCL @arxiv_cscl@social.cleverlibre.org

Alternatives to the Scaled Dot Product for Attention in the Transformer Neural Network Architecture. (arXiv:2311.09406v1 [cs.LG])

http://arxiv.org/abs/2311.09406 #arXiv #NLProc

Nov 19, 2023, 03:18 · · arxiv-cscl · · ·

Sign in to participate in the conversation