Improving language models by retrieving from trillions of tokens.
Sebastian BorgeaudArthur MenschJordan HoffmannTrevor CaiEliza RutherfordKatie MillicanGeorge van den DriesscheJean-Baptiste LespiauBogdan DamocAidan ClarkDiego de Las CasasAurelia GuyJacob MenickRoman RingTom HenniganSaffron HuangLoren MaggioreChris JonesAlbin CassirerAndy BrockMichela PaganiniGeoffrey IrvingOriol VinyalsSimon OsinderoKaren SimonyanJack W. RaeErich ElsenLaurent SifrePublished in: CoRR (2021)
Keyphrases
- language model
- information retrieval
- language modeling
- n gram
- speech recognition
- probabilistic model
- document retrieval
- language modelling
- retrieval model
- context sensitive
- ad hoc information retrieval
- test collection
- statistical language models
- vector space model
- query expansion
- relevance model
- translation model
- information retrieval systems
- language models for information retrieval
- word error rate
- statistical language modeling
- language model for information retrieval
- document length
- document ranking
- visual features
- keywords
- search engine