Frequency Explains the Inverse Correlation of Large Language Models' Size, Training Data Amount, and Surprisal's Fit to Reading Times.
Byung-Doh OhShisen YueWilliam SchulerPublished in: EACL (1) (2024)
Keyphrases
- language model
- training data
- language modeling
- n gram
- document retrieval
- information retrieval
- statistical language models
- probabilistic model
- speech recognition
- language modelling
- query expansion
- retrieval model
- ad hoc information retrieval
- language model for information retrieval
- smoothing methods
- test collection
- relevance model
- translation model
- learning algorithm
- pseudo relevance feedback
- context sensitive
- document length
- training set
- language models for information retrieval
- expert search
- term dependencies
- vector space model