arXiv stat.ML
· Papers
Self-Organized Conformal Prediction: Reducing Regional Coverage Gaps with Unsupervised Group Discovery
arXiv:2606.29403v1 Announce Type: new Abstract: Conformal prediction guarantees marginal coverage, but pooled calibration averages over heterogeneous regions and can mask regional undercoverage in safety-critical subgroups. We introduce Self-Organized Conformal Prediction (SOCP), a calibration scheme that discovers inp