Findings of the 2023 Conference on Machine Translation (WMT23): LLMs Are Here But Not Quite There Yet - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2023

Findings of the 2023 Conference on Machine Translation (WMT23): LLMs Are Here But Not Quite There Yet

Tom Kocmi
  • Function : Author
Anton Dvorkovich
  • Function : Author
Mark Fishel
  • Function : Author
Benjamin Marie
  • Function : Author
Mariya Shmatova
  • Function : Author

Abstract

This paper presents the results of the General Machine Translation Task organised as part of the 2023 Conference on Machine Translation (WMT). In the general MT task, participants were asked to build machine translation systems for any of 8 language pairs (covering 14 translation directions), to be evaluated on test sets consisting of up to four different domains. We evaluate system outputs with professional human annotators using a combination of source-based Direct Assessment and scalar quality metric (DA+SQM).
Fichier principal
Vignette du fichier
Findings_of_the_2023_Conference_on_Machine_Translation__WMT23_.pdf (637.65 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-04300702 , version 1 (22-11-2023)

Licence

Attribution

Identifiers

  • HAL Id : hal-04300702 , version 1

Cite

Tom Kocmi, Eleftherios Avramidis, Rachel Bawden, Ondřej Bojar, Anton Dvorkovich, et al.. Findings of the 2023 Conference on Machine Translation (WMT23): LLMs Are Here But Not Quite There Yet. WMT23 - Eighth Conference on Machine Translation, Dec 2023, Singapore, Singapore. pp.198--216. ⟨hal-04300702⟩
96 View
100 Download

Share

Gmail Facebook X LinkedIn More