Overlap-based Vocabulary Generation Improves Cross-lingual Transfer Among Related Languages.
Vaidehi PatilPartha P. TalukdarSunita SarawagiPublished in: ACL (1) (2022)
Keyphrases
- cross lingual
- language independent
- transfer learning
- cross lingual information retrieval
- machine translation
- multi lingual
- language modeling
- cross language
- event extraction
- european languages
- parallel corpora
- out of vocabulary
- language specific
- text classification
- indian languages
- word segmentation
- query translation
- word alignment
- machine learning
- mono lingual
- target language
- linguistic resources
- comparable corpora
- active learning
- statistical machine translation
- news articles
- document clustering
- n gram
- generative model
- probabilistic model
- search engine
- learning algorithm