Your fairness may vary: Pretrained language model fairness in toxic text classification.
Ioana BaldiniDennis WeiKarthikeyan Natesan RamamurthyMoninder SinghMikhail YurochkinPublished in: ACL (Findings) (2022)
Keyphrases
- language model
- language modeling
- text classification
- n gram
- probabilistic model
- document retrieval
- language modelling
- retrieval model
- speech recognition
- information retrieval
- query expansion
- test collection
- language model for information retrieval
- statistical language models
- machine learning
- document ranking
- query terms
- ad hoc information retrieval
- relevance model
- context sensitive
- bag of words
- cross lingual
- vector space model
- document length
- feature selection
- statistical language modeling
- language models for information retrieval
- mixture model
- word clouds
- data mining
- text classifiers
- smoothing methods
- term frequency
- knn
- text mining
- text documents
- multi label