BERTSeg: BERT Based Unsupervised Subword Segmentation for Neural Machine Translation.
Haiyue SongRaj DabreZhuoyuan MaoChenhui ChuSadao KurohashiPublished in: AACL/IJCNLP (2) (2022)
Keyphrases
- machine translation
- cross lingual
- pos tagging
- natural language processing
- language processing
- information extraction
- language independent
- target language
- chinese english
- word alignment
- cross language information retrieval
- brazilian portuguese
- machine translation system
- natural language
- word sense disambiguation
- natural language generation
- statistical machine translation
- language resources
- word segmentation
- grammar induction
- query translation
- n gram
- semi supervised
- image segmentation
- co occurrence
- lexical semantics
- machine transliteration
- knowledge base