Multiple-Debias reduces gender, racial, and religious biases in multilingual pre-trained language models more effectively than monolingual methods by integrating counterfactual augmentation and self-debiasing across pre- and post-processing stages in four languages.
Gender bias in masked language models for multiple languages
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Multiple-Debias: A Full-process Debiasing Method for Multilingual Pre-trained Language Models
Multiple-Debias reduces gender, racial, and religious biases in multilingual pre-trained language models more effectively than monolingual methods by integrating counterfactual augmentation and self-debiasing across pre- and post-processing stages in four languages.