Lexical impoverishment in neural machine translation
The information required to translate a word in the machine translation of isolated sentences is often missing or unobtainable. We show that the diversity in the training-set distribution of Spanish equivalents of English nouns is reduced in translation. A translator survey reveals this is often due...
| Autores: | , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2024 |
| País: | España |
| Institución: | Universitat Autònoma de Barcelona |
| Repositorio: | Dipòsit Digital de Documents de la UAB |
| Idioma: | inglés |
| OAI Identifier: | oai:ddd.uab.cat:305589 |
| Acceso en línea: | https://ddd.uab.cat/record/305589 https://dx.doi.org/urn:doi:10.5565/rev/tradumatica.428 |
| Access Level: | acceso abierto |
| Palabra clave: | Machine translation Spanish-english Lexical empoverishment Lexical equivalent Distribution distortion Traducció automàtica Espanyol-anglès Empobriment lèxic Equivalent lèxic Distorsió de la distribució Traducción automática Español-inglés Empobrecimiento léxico Equivalente léxico Distorsión de la distribución |
| Sumario: | The information required to translate a word in the machine translation of isolated sentences is often missing or unobtainable. We show that the diversity in the training-set distribution of Spanish equivalents of English nouns is reduced in translation. A translator survey reveals this is often due to lack of source context. |
|---|