SelfSeg: A Self-supervised Sub-word Segmentation Method for Neural Machine Translation.
Haiyue SongRaj DabreChenhui ChuSadao KurohashiEiichiro SumitaPublished in: CoRR (2023)
Keyphrases
- segmentation method
- machine translation
- word sense disambiguation
- statistical machine translation
- word level
- word alignment
- target language
- machine translation system
- parallel corpus
- source language
- english chinese
- statistical translation models
- language independent
- image segmentation
- language processing
- bilingual dictionaries
- tasks in natural language processing
- target word
- active contours
- cross lingual
- information extraction
- natural language processing
- natural language
- cross language information retrieval
- chinese english
- grammar induction
- co occurrence
- segmentation algorithm
- bilingual lexicon
- parallel corpora
- language resources
- segmented images
- word segmentation
- word recognition
- natural language generation
- word order
- conditional random fields
- energy function
- n gram
- machine learning
- part of speech
- knowledge base
- feature selection
- artificial intelligence