Towards Personalised and Document-level Machine Translation of Dialogue

Sebastian Vincent

Student Research Workshop Long paper Paper

Zoom-5E: Apr 22, Zoom-5E: Apr 22 (12:00-13:00 UTC) [Join Zoom Meeting]
Gather-2F: Apr 22, Gather-2F: Apr 22 (13:00-15:00 UTC) [Join Gather Meeting]

Abstract: State-of-the-art (SOTA) neural machine translation (NMT) systems translate texts at sentence level, ignoring context: intra-textual information, like the previous sentence, and extra-textual information, like the gender of the speaker. As a result, some sentences are translated incorrectly. Personalised NMT (PersNMT) and document-level NMT (DocNMT) incorporate this information into the translation process. Both fields are relatively new and previous work within them is limited. Moreover, there are no readily available robust evaluation metrics for them, which makes it difficult to develop better systems, as well as track global progress and compare different methods. This thesis proposal focuses on PersNMT and DocNMT for the domain of dialogue extracted from TV subtitles in five languages: English, Brazilian Portuguese, German, French and Polish. Three main challenges are addressed: (1) incorporating extra-textual information directly into NMT systems; (2) improving the machine translation of cohesion devices; (3) reliable evaluation for PersNMT and DocNMT.

Connected Papers in EACL2021

Similar Papers