LICHEE: Improving Language Model Pre-training with Multi-grained Tokenization.
Weidong GuoMingjun ZhaoLusheng ZhangDi NiuJinwen LuoZhenhua LiuZhenyang LiJianbo TangPublished in: ACL/IJCNLP (Findings) (2021)
Keyphrases
- language model
- n gram
- language modeling
- document retrieval
- probabilistic model
- information retrieval
- language modelling
- query expansion
- speech recognition
- pseudo relevance feedback
- retrieval model
- context sensitive
- mixture model
- statistical language models
- named entities
- vector space model
- smoothing methods
- relevance model
- document ranking
- test collection
- ad hoc information retrieval
- translation model
- query terms
- information extraction
- cross language retrieval
- word error rate
- dependency structure
- supervised learning
- unsupervised learning
- language modeling framework
- language models for information retrieval