What does it take to create the Babel Fish, a tool that can help individ...
We introduce SONAR, a new multilingual and multimodal fixed-size sentenc...
We introduce a new proxy score for evaluating bitext mining based on
sim...
End-to-End speech-to-speech translation (S2ST) is generally evaluated wi...
We study speech-to-speech translation (S2ST) that translates speech from...
We present SpeechMatrix, a large-scale multilingual corpus of
speech-to-...
Image generation has recently seen tremendous advances, with diffusion m...
Multilingual sentence representations from large models can encode seman...
Driven by the goal of eradicating language barriers on a global scale,
m...
Scaling multilingual representation learning beyond the hundred most fre...
We present a new approach to perform zero-shot cross-modal transfer betw...
Deep generative models, like GANs, have considerably improved the state ...
We present a textless speech-to-speech translation (S2ST) system that ca...
Latent text representations exhibit geometric regularities, such as the
...
In this paper, we describe our end-to-end multilingual speech translatio...
Existing work in translation demonstrated the potential of massively
mul...
We show that margin-based bitext mining in a multilingual sentence space...
Question answering (QA) models have shown rapid progress enabled by the
...
We present an approach based on multilingual sentence embeddings to
auto...
In this paper, we describe our submission to the WMT19 low-resource para...
We introduce an architecture to learn joint multilingual sentence
repres...
Machine translation is highly sensitive to the size and quality of the
t...
State-of-the-art natural language processing systems rely on supervision...
We learn a joint multilingual sentence embedding and use the distance be...
Cross-lingual document classification aims at training a document classi...
Many modern NLP systems rely on word embeddings, previously trained in a...
In this paper, we use the framework of neural machine translation to lea...
The dominant approach for many NLP tasks are recurrent neural networks, ...
Recent work on end-to-end neural network-based architectures for machine...
It is today acknowledged that neural network language models outperform
...
In this paper, we propose a novel neural network model called RNN
Encode...