MOAI speeds non-interactive homomorphic transformer inference
It cuts BERT-base evaluation time 52.8% versus THOR and reports 2.36 minutes amortized per input on one GPU.
Stories per day · 30 days
1 stories, newest first · one card per story
1 stories, newest first · one card per story
It cuts BERT-base evaluation time 52.8% versus THOR and reports 2.36 minutes amortized per input on one GPU.
Today 707 readers · 836 views · 275 reads · all time 187K views