PiCkLe at SemEval-2022 Task 4: Boosting Pre-trained Language Models with Task Specific Metadata and Cost Sensitive Learning.
Manan SuriPublished in: SemEval@NAACL (2022)
Keyphrases
- language model
- cost sensitive learning
- pre trained
- cost sensitive
- misclassification costs
- training data
- training examples
- information retrieval
- document retrieval
- probabilistic model
- speech recognition
- multi class
- active learning
- class imbalance
- missing values
- naive bayes
- test collection
- multimedia
- base classifiers
- class distribution
- machine learning
- learning algorithm
- support vector machine
- neural network
- generative model
- image classification
- classification accuracy
- decision trees
- multi label
- em algorithm
- e learning