cushLEPOR: customising hLEPOR metric using Optuna for higher agreement with human judgments or pre-trained language model LaBSE.
Lifeng HanIrina SorokinaGleb ErofeevSerge GladkoffPublished in: WMT@EMNLP (2021)
Keyphrases
- language model
- pre trained
- human judgments
- language modeling
- n gram
- document retrieval
- retrieval model
- information retrieval
- probabilistic model
- query expansion
- speech recognition
- semantic relatedness
- training examples
- training data
- test collection
- document representation
- distance measure
- feature selection
- training set
- learning algorithm