A big convergence of language, multimodal perception, action, and world
...
In this paper, we elaborate upon recipes for building multilingual
repre...
A big convergence of model architectures across language, vision, speech...
A big convergence of language, vision, and multimodal pretraining is
eme...
Sparse mixture of experts provides larger model capacity while requiring...
This report describes Microsoft's machine translation systems for the WM...
Compared to monolingual models, cross-lingual models usually require a m...
In this paper, we introduce ELECTRA-style tasks to cross-lingual languag...
While pretrained encoders have achieved success in various natural langu...
Fine-tuning pre-trained cross-lingual language models can transfer
task-...
Multilingual machine translation enables a single model to translate bet...
In this work, we formulate cross-lingual language model pre-training as
...