OxEnsemble: Fair Ensembles for Low-Data Classification

arXiv:2512.09665v2 Announce Type: replace Abstract: We address the problem of fair classification in settings where data is scarce and unbalanced across demographic groups. Such low-data regimes are common in domains like medical imaging, where false negatives can have fatal consequences. We propose a novel approach OxEnsemble for efficiently training ensembles and enforcing fairness in these low-data regimes. Unlike other approaches, we aggregate predictions across ensemble members, each trained to satisfy fairness constraints. By construction, OxEnsemble is both data-efficient -- carefully reusing held-out data to enforce fairness reliably -- and compute-efficient, requiring little more compute than used to fine-tune or evaluate an existing model. We validate this approach with new theoretical guarantees. Experimentally, our approach yields more consistent outcomes and stronger fairness-accuracy trade-offs than existing methods across multiple challenging medical imaging classification datasets.

Related

• Bias Redistribution in Visual Machine Unlearning: Does Forgetting One Group Harm Another?
• SeMoBridge: Semantic Modality Bridge for Efficient Few-Shot Adaptation of CLIP
• Shortcut Learning in Glomerular AI: Adversarial Penalties Hurt, Entropy Helps
• Inside-Out: Measuring Generalization in Vision Transformers Through Inner Workings

Source: arXiv cs.CV | 2026-04-10