The success of Transformer architecture has seen increased interest in machine translation
(MT). The translation quality of neural network-based MT transcends that of translations …

The ability of zero-shot translation emerges when we train a multilingual model with certain
translation directions; the model can then directly translate in unseen directions …

This paper describes the AST submission to the WMT23 Shared Task on Parallel Data
Curation. We experiment with two approaches for curating data from the provided web …

Lexical simplification (LS) methods based on pretrained language models have made
remarkable progress, generating potential substitutes for a complex word through analysis …

When parallel corpora are preprocessed for machine translation (MT) training, a part of the
parallel data is commonly discarded and deemed non-parallel due to odd-length ratio …

For machine translation (MT) systems to produce accurate and fluent translations, reliable
parallel corpora are key. Errors, due to misalignments or inadequate filtering during …

This thesis describes an unsupervised approach to determine the translation direction for
parallel texts. Traditional methods in this field rely on large amounts of homogeneous …

Jannis Vamvas - My PhD Thesis Is Out! (A Summary) My PhD Thesis Is Out! (A Summary) April
05, 2023 Avatar Jannis Vamvas NLP Researcher, University of Zurich After quite a few months …