TAeKD: Teacher Assistant Enhanced Knowledge Distillation for Closed-Source Multilingual Neural Machine Translation.
Bo LvXin LiuKaiwen WeiPing LuoYue YuPublished in: LREC/COLING (2024)
Keyphrases
- machine translation
- cross lingual
- language independent
- cross language information retrieval
- language resources
- target language
- chinese english
- multilingual documents
- parallel corpus
- information extraction
- machine translation system
- natural language processing
- word sense disambiguation
- statistical machine translation
- language processing
- source language
- natural language generation
- lexical knowledge
- cross lingual information retrieval
- word alignment
- language specific
- natural language
- cross language
- multilingual information retrieval
- tasks in natural language processing
- co occurrence
- knowledge representation
- knowledge base
- comparable corpora
- parallel corpora
- word level
- brazilian portuguese
- data mining