CokeBERT: Contextual knowledge selection and embedding towards enhanced pre-trained language models.
Yusheng SuXu HanZhengyan ZhangYankai LinPeng LiZhiyuan LiuJie ZhouMaosong SunPublished in: AI Open (2021)
Keyphrases
- language model
- contextual knowledge
- pre trained
- language modeling
- contextual information
- probabilistic model
- document retrieval
- n gram
- case based reasoning
- information retrieval
- query expansion
- retrieval model
- training data
- query terms
- speech recognition
- test collection
- machine learning
- vector space
- language models for information retrieval
- context aware
- retrieval effectiveness
- domain specific
- relevance model
- hidden markov models
- smoothing methods
- feature extraction