BeliefBank: Adding Memory to a Pre-Trained Language Model for a Systematic Notion of Belief.
Nora KassnerOyvind TafjordHinrich SchützePeter ClarkPublished in: EMNLP (1) (2021)
Keyphrases
- language model
- pre trained
- language modeling
- n gram
- document retrieval
- probabilistic model
- training data
- speech recognition
- language modelling
- retrieval model
- information retrieval
- training examples
- test collection
- context sensitive
- mixture model
- query expansion
- control signals
- ad hoc information retrieval
- smoothing methods
- relevance model
- translation model
- data sets
- training set
- feature space
- information extraction
- cross lingual
- word clouds