BERTIN: Efficient Pre-Training of a Spanish Language Model using Perplexity Sampling.
Javier de la RosaEduardo G. PonferradaManu RomeroPaulo VillegasPablo González de Prado SalasMaría GranduryPublished in: Proces. del Leng. Natural (2022)
Keyphrases
- language model
- language modeling
- n gram
- speech recognition
- document retrieval
- probabilistic model
- information retrieval
- mixture model
- language modelling
- query expansion
- word error rate
- test collection
- retrieval model
- context sensitive
- statistical language models
- pseudo relevance feedback
- relevance model
- co occurrence
- training set
- vector space model
- translation model
- multiword
- document ranking
- smoothing methods
- cross language retrieval
- search engine