MiLMo: Minority Multilingual Pre-trained Language Model.
Hanru ShiSisi LiuXinhe YuWugedele BaoYuan SunXiaobing ZhaoPublished in: CoRR (2022)
Keyphrases
- language model
- pre trained
- language modeling
- cross lingual
- training data
- document retrieval
- n gram
- cross language
- training examples
- speech recognition
- probabilistic model
- information retrieval
- cross language information retrieval
- query terms
- query expansion
- mixture model
- test collection
- retrieval model
- control signals
- context sensitive
- smoothing methods
- ad hoc information retrieval
- supervised learning
- small number
- support vector
- relevance model
- translation model
- dirichlet prior