Direction is what you need: Improving Word Embedding Compression in Large Language Models.
Klaudia BalazyMohammadreza BanaeiRémi LebretJacek TaborKarl AbererPublished in: RepL4NLP@ACL-IJCNLP (2021)
Keyphrases
- language model
- n gram
- language modeling
- translation model
- word error rate
- statistical language modeling
- probabilistic model
- information retrieval
- word clouds
- query expansion
- document retrieval
- language modelling
- out of vocabulary
- retrieval model
- speech recognition
- multiword
- test collection
- ad hoc information retrieval
- statistical language models
- smoothing methods
- spoken term detection
- language independent
- context sensitive
- vector space model
- document ranking
- word segmentation
- term weighting
- vector space
- relevance model
- co occurrence