Mitigating Long-Tail Language Representation Collapsing via Cross-Lingual Bootstrapped Unsupervised Fine-Tuning.
Ping GuoYue HuYubing RenYunpeng LiJiarui ZhangXingsheng ZhangPublished in: ECAI (2023)
Keyphrases
- fine tuning
- cross lingual
- long tail
- parallel corpus
- machine translation
- linguistic resources
- language independent
- source language
- cross lingual information retrieval
- indian languages
- language modeling
- cross language
- machine translation system
- bilingual dictionaries
- target language
- text classification
- parallel corpora
- supervised learning
- natural language
- query translation
- statistical machine translation
- language model