Fight Fire with Fire: Fine-tuning Hate Detectors using Large Samples of Generated Hate Speech.
Tomer WullachAmir AdlerEinat MinkovPublished in: EMNLP (Findings) (2021)
Keyphrases
- fine tuning
- fine tune
- viable alternative
- data sets
- fine tuned
- automatically generated
- object detection
- speech recognition
- audio visual
- speech synthesis
- spectral features
- language acquisition
- speech signal
- training samples
- pattern recognition
- computer vision
- automatic speech recognition
- user interface
- generation method
- broadcast news
- multiscale
- learning algorithm
- machine learning
- neural network