Existing work in translation demonstrated the potential of massively multilingual machine
translation by training a single model able to translate between any pair of languages …

Probabilistic topic models are unsupervised generative models which model document
content as a two-step generation process, that is, documents are observed as mixtures of …

We present an approach based on multilingual sentence embeddings to automatically
extract parallel sentences from the content of Wikipedia articles in 85 languages, including …

We report on methods to create the largest publicly available parallel corpora by crawling
the web, using open source software. We empirically compare alternative methods and …

This book provides an overview of various techniques for the alignment of bitexts. It
describes general concepts and strategies that can be applied to map corresponding parts …

Neural machine translation (NMT) is a deep learning based approach for machine
translation, which outperforms traditional statistical machine translation (SMT) and yields the …

We show that margin-based bitext mining in a multilingual sentence space can be applied to
monolingual corpora of billions of sentences. We are using ten snapshots of a curated …

Machine translation is highly sensitive to the size and quality of the training data, which has
led to an increasing interest in collecting and filtering large parallel corpora. In this paper, we …

We control the world basically because we are the only animals that can cooperate flexibly
in very large numbers […] This is something very unique to us, perhaps the most unique …

In this paper, we present an approach to learn multilingual sentence embeddings using a bi-
directional dual-encoder with additive margin softmax. The embeddings are able to achieve …