ModelMultilingual BERT
M-BERT
A multilingual pretrained version of BERT, jointly trained on corpora covering 104 languages with a shared vocabulary and shared parameters, supporting cross-lingual text understanding.
Timeline (last 90 days)
Oct 25
Laya AI uses a BERT/M-BERT backbone network to score candidate answers discriminatively, rather than generating text token by token.
Unverified50%
Oct 25
M-BERT is a multilingual version of BERT jointly trained on Wikipedia corpora in 104 languages.
Unverified50%