WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models.
Benjamin MinixhoferFabian PaischerNavid RekabsazPublished in: NAACL-HLT (2022)
Keyphrases
- cross lingual
- language modeling
- language model
- n gram
- translation model
- pseudo feedback
- speech recognition
- cross language retrieval
- retrieval model
- query expansion
- out of vocabulary
- cross lingual information retrieval
- language independent
- cross language
- information retrieval
- document retrieval
- language modeling framework
- probabilistic model
- machine translation
- text classification
- test collection
- transfer learning
- parallel corpus
- context sensitive
- parallel corpora
- pseudo relevance feedback
- smoothing methods
- word segmentation
- word alignment
- machine translation system
- feature selection
- text mining
- bilingual dictionaries
- statistical machine translation
- relevance model
- query translation
- vector space
- query terms
- retrieval effectiveness