Stereotype and Skew: Quantifying Gender Bias in Pre-trained and Fine-tuned Language Models.
Daniel de Vassimon ManelaDavid ErringtonThomas FisherBoris van BreugelPasquale MinerviniPublished in: EACL (2021)
Keyphrases
- language model
- pre trained
- fine tuned
- fine tuning
- language modeling
- training data
- query expansion
- document retrieval
- n gram
- information retrieval
- speech recognition
- retrieval model
- training examples
- probabilistic model
- control signals
- test collection
- language modelling
- domain specific
- smoothing methods
- statistical language models
- prior knowledge
- spoken term detection
- machine learning
- multimedia
- information extraction
- small number
- human body
- dimensionality reduction