Hyperparameter Power Impact in Transformer Language Model Training.
Lucas Høyberg Puvis de ChavannesMads Guldborg Kjeldgaard KongsbakTimmie RantzauLeon DerczynskiPublished in: SustaiNLP@EMNLP (2021)
Keyphrases
- language model
- language modeling
- document retrieval
- n gram
- probabilistic model
- speech recognition
- information retrieval
- context sensitive
- query expansion
- statistical language models
- test collection
- retrieval model
- language modelling
- language model for information retrieval
- ad hoc information retrieval
- vector space model
- document ranking
- smoothing methods
- pseudo relevance feedback
- relevance model
- language models for information retrieval
- training set
- retrieval effectiveness
- word error rate
- co occurrence
- knn