We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.CL

Change to browse by:

cs

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computation and Language

Title: Tilde at WMT 2020: News Task Systems

Abstract: This paper describes Tilde's submission to the WMT2020 shared task on news translation for both directions of the English-Polish language pair in both the constrained and the unconstrained tracks. We follow our submissions from the previous years and build our baseline systems to be morphologically motivated sub-word unit-based Transformer base models that we train using the Marian machine translation toolkit. Additionally, we experiment with different parallel and monolingual data selection schemes, as well as sampled back-translation. Our final models are ensembles of Transformer base and Transformer big models that feature right-to-left re-ranking.
Subjects: Computation and Language (cs.CL)
Cite as: arXiv:2010.15423 [cs.CL]
  (or arXiv:2010.15423v1 [cs.CL] for this version)

Submission history

From: Mārcis Pinnis [view email]
[v1] Thu, 29 Oct 2020 08:59:37 GMT (7107kb)

Link back to: arXiv, form interface, contact.