{
  "id": 7269,
  "url": "https://arxiv.org/abs/2603.13452v3",
  "title": "MESD: A Risk-Sensitive Metric for Explanation Fairness Across Intersectional Subgroups",
  "summary": "Fairness in machine learning is predominantly evaluated through outcome-oriented metrics, such as Demographic parity, which measure whether predictions are statistically consistent across protected groups. However, these metrics cannot detect whether a model uses systematically different reasoning for different demographic groups, which violates procedural fairness principles. This problem is compounded by intersectionality, where models may appear fair on individual attributes (e.g., race) whil",
  "authors": "Gideon Popoola, John Sheppard",
  "category": "research",
  "topics": "bias-fairness",
  "orgs": null,
  "regions": null,
  "published_at": "2026-03-13T15:42:31.000Z",
  "fetched_at": "2026-07-14T16:33:08.011Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/7269",
  "original_url": "https://arxiv.org/abs/2603.13452v3",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}