Simulated multiple reference training improves low-resource machine translation.
Huda KhayrallahBrian ThompsonMatt PostPhilipp KoehnPublished in: EMNLP (1) (2020)
Keyphrases
- machine translation
- information extraction
- natural language processing
- target language
- language independent
- statistical machine translation
- cross lingual
- language processing
- cross language information retrieval
- natural language
- word alignment
- natural language generation
- brazilian portuguese
- word level
- artificial intelligence
- source language
- multilingual documents
- machine translation system
- knowledge base
- query translation
- cross language
- statistical model
- parallel corpus
- language resources
- machine learning