Training Hybrid Language Models by Marginalizing over Segmentations.
Edouard GraveSainbayar SukhbaatarPiotr BojanowskiArmand JoulinPublished in: ACL (1) (2019)
Keyphrases
- language model
- language modeling
- document retrieval
- n gram
- retrieval model
- query expansion
- information retrieval
- probabilistic model
- speech recognition
- test collection
- language modelling
- statistical language models
- smoothing methods
- context sensitive
- language model for information retrieval
- ad hoc information retrieval
- language models for information retrieval
- document ranking
- word error rate
- training set
- image segmentation
- pseudo relevance feedback
- relevance model
- document length
- retrieval effectiveness
- cross lingual
- translation model
- text retrieval
- query terms
- okapi bm