How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models.
Phillip RustJonas PfeifferIvan VulicSebastian RuderIryna GurevychPublished in: ACL/IJCNLP (1) (2021)
Keyphrases
- language model
- cross lingual
- language modeling
- cross language
- document retrieval
- query expansion
- multilingual information retrieval
- cross lingual information retrieval
- cross language retrieval
- language independent
- n gram
- translation model
- information retrieval
- cross language information retrieval
- query specific
- probabilistic model
- parallel corpus
- chinese english
- retrieval model
- speech recognition
- query terms
- statistical machine translation
- statistical language models
- context sensitive
- language modelling
- text retrieval
- test collection
- pseudo feedback
- machine translation system
- parallel corpora
- multilingual retrieval
- passage retrieval
- question answering
- relevance model
- retrieval effectiveness
- vector space model
- text categorization
- document collections
- machine translation
- information access
- smoothing methods
- query translation
- pseudo relevance feedback
- bilingual dictionaries
- out of vocabulary
- source language
- topic models
- document ranking
- word segmentation
- ad hoc retrieval