Adapting High-resource NMT Models to Translate Low-resource Related Languages without Parallel Data.
Wei-Jen KoAhmed El-KishkyAdithya RenduchintalaVishrav ChaudharyNaman GoyalFrancisco GuzmánPascale FungPhilipp KoehnMona T. DiabPublished in: ACL/IJCNLP (1) (2021)
Keyphrases
- data sets
- experimental data
- data quality
- input data
- databases
- high correlation
- statistical methods
- data structure
- probabilistic model
- high dimensional data
- synthetic data
- database
- data mining techniques
- data points
- training data
- information resources
- historical data
- raw data
- learned models
- resource allocation
- accurate models
- data analysis
- machine learning
- statistical analysis
- data collection
- data sources
- resource usage
- data transfer
- language independent
- cost benefit analysis
- parallel processing
- decision trees
- original data
- data distribution
- data streams
- spatial data
- prior knowledge
- probability distribution
- knowledge discovery