search

zehuiwu / MMML

Multi-Modality Multi-Loss Fusion Network
43 stars 7 forks source link

CME Layer #2

Closed M-yanan closed 2 months ago

M-yanan commented 3 months ago

Hello, may I ask if you don't use the cross attention layer when adding context modeling? Why not use them together?

gravity5212 commented 3 months ago

I also have the same question

zehuiwu commented 2 months ago

updated!