method · first seen 15 Oct 2026
adversarial training for fairness
Also: adversarial regularization
Training a model with an adversary that removes demographic information from representations to reduce bias.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Mitigating Demographic Disparities in Explanation-Human Alignment via Rationale RegularizationPapers · 15 Oct 2026 · Suryansh Talukdar, Taha Sarwat, Jonathan Rusert
Latest sources
- Mitigating Demographic Disparities in Explanation-Human Alignment via Rationale RegularizationPapers · 15 Oct 2026 · Suryansh Talukdar, Taha Sarwat, Jonathan Rusert