Text encoders are performance bottlenecks in contrastive vision-language models.
Amita KamathJack HesselKai-Wei ChangPublished in: CoRR (2023)
Keyphrases
- language model
- information retrieval
- language modeling
- n gram
- document level
- retrieval model
- document retrieval
- speech recognition
- text retrieval
- query expansion
- probabilistic model
- language modelling
- vector space model
- test collection
- text mining
- web documents
- statistical language models
- smoothing methods
- term dependencies
- multiword
- relevance model
- pseudo relevance feedback
- ad hoc information retrieval
- language models for information retrieval
- document ranking
- okapi bm
- context sensitive
- language model for information retrieval
- statistical language modeling
- word level
- speech signal
- relevance feedback
- keywords
- bayesian networks