The Devil is in the Details: On Models and Training Regimes for Few-Shot Intent Classification.

Mohsen Mesgar Thy Thy Tran Goran Glavas Iryna Gurevych

Published in: EACL (2023)

Keyphrases

classification models
supervised learning
classification accuracy
machine learning
machine learning algorithms
machine learning methods
model selection
support vector machine

training set
training process
training phase
feature selection
neural network
statistical models
classification scheme
training dataset

automatic classification
classification performances
training samples
test set
unsupervised learning
feature set
feature extraction
classification algorithm

classification method
support vector machine svm
text classification
image classification
feature vectors
feature space
decision trees
tangent distance