Users’ needs in interactive bias auditing tools introducing a requirement checklist and evaluating existing tools
Source
AI and ethics - ISSN 2730-5953-5 (2025) p. 341-369
FairFlow : an automated approach to model-based counterfactual data augmentation for NLP
Source
Lecture notes in computer science-14947 (2024) p. 160-176
Interpretable and fair mechanisms for abstaining classifiers
Source
Lecture notes in computer science-14947 (2024) p. 416-433
Assessing and mitigating bias in natural language systems
Source
Antwerpen, University of Antwerp, Faculty of Science, 2024,xvi, 136 p.
Targeted approaches against discrimination : new methods for bias detection and mitigation in automated decision-making systems
Source
Antwerpen, University of Antwerp, Faculty of Science, 2024,180 p.