这论文用海量实验证明，碰到严重不平衡数据时，传统预测区间会漏掉少数类，Mondrian CP加弃权能大幅提升覆盖率和成本效益，做高风险决策的值得看。
针对信用评分、欺诈检测等高风险中类别不平衡和不对称错误成本问题，该研究比较了边际共形预测（marginal CP）、类条件共形预测（Mondrian CP）和成本控制弃权机制。在15个真实不平衡数据集、7个分类模型、3种校准方法和10个随机种子下进行3150次实验。结果显示Mondrian CP将少数类覆盖率平均提升61.7个百分点（p<1e-80），而边际CP在某些数据集上少数类覆盖率低至0.5%。结合成本控制弃权还进一步降低了预期决策成本，并量化了将模糊实例交给人类专家判断的盈亏平衡阈值。
Cost-Sensitive Conformal Prediction and Human-in-the-Loop Abstention for Imbalanced High-Stakes Decision Support: A Multi-Domain Benchmark
High-stakes decision systems in credit scoring, fraud detection, healthcare, and industrial safety require reliable uncertainty quantification under severe class imbalance and asymmetric error costs. Standard marginal conformal prediction (CP) provides valid overall coverage guarantees; however, we show that it severely under-covers rare, costly minority classes, with minority-class coverage dropping to as low as 0.5% on certain datasets. To characterize and address this limitation, we conduct a comprehensive benchmark comparing marginal CP, class-conditional (Mondrian) CP, and cost-controlled abstention mechanisms across 15 real-world imbalanced tabular datasets, 7 classification models, 3 probability calibration techniques, and 10 random seeds, resulting in 3,150 experimental runs. Our results show that Mondrian CP restores valid minority-class coverage, achieving an average minority-coverage improvement of 61.7 percentage points over marginal CP (p < 1e-80). Furthermore, combining Mondrian CP with cost-controlled abstention significantly reduces expected decision cost compared with standard decision boundaries, confidence-based rejectors, and risk-controlled rejectors under realistic human review budgets. We further quantify dataset-specific break-even thresholds at which deferring ambiguous instances to human experts becomes cost-effective. These findings provide practical guidance for deploying distribution-free, cost-aware uncertainty quantification in high-stakes decision support systems.