Login / Signup

Don't Discard All the Biased Instances: Investigating a Core Assumption in Dataset Bias Mitigation Techniques.

Hossein AmirkhaniMohammad Taher Pilehvar
Published in: EMNLP (Findings) (2021)
Keyphrases
  • randomly generated
  • benchmark datasets
  • neural network
  • real life
  • multi class
  • text classification
  • training instances