Universal Language Model Fine-Tuning (ULMfit)

Comparining Universal Language Model Fine-Tuning (ULMfit) with other traditional approaches

  • This method uses single training and architecture.
  • This method does not require labels or in-domain documents.

Universal Language Model Fine-Tuning Usages

  • In a project where the tasks vary according to the size of documents, labels, and numbers.
  • In a task where there is no requirement of pre-processing and feature engineering.