针对空间相关数据的评估偏差，提出结构感知分层划分与课程鲁棒优化方法。
Beyond the Performance Illusion: Structure-Aware Stratified Partitioning and Curriculum Distributionally Robust Optimization for Spatially Correlated Domains
- 基于空间结构设计分层划分策略，减少数据泄露风险。
- 在多个基准上提升泛化性能，暴露传统评估隐藏的失败模式。
- 适合遥感、医疗影像等空间相关领域的模型评估与训练。
AI系统性能评估通常假设随机数据划分产生独立同分布子集。我们发现，在航拍监控、精准农业和医学影像等时空相关领域，该假设常失效，导致两大系统性问题：数据泄露（相关样本跨越训练与验证集，虚高性能估计）和隐藏分层（少数子群体错误被整体指标掩盖）。为此，我们提出统一的评估与训练框架：引入结构感知分层划分（SASP），构建减少时空泄漏且保持类别平衡的验证集；提出课程分布鲁棒优化（CDRO），通过课程式松弛实现分布鲁棒训练的稳定优化。在多个基准测试中，该组合显著提升泛化能力，改善置信度校准，并揭示传统随机划分下隐藏的模型失效模式。
原文摘要 · Abstract (English)
Performance evaluation in AI systems commonly assumes that random dataset splits produce independent and identically distributed (i.i.d.) subsets. We show that this assumption often breaks down in spatiotemporally correlated domains such as aerial surveillance, precision agriculture, and medical imaging, leading to two systematic failures: data leakage, where correlated samples span training and validation splits and inflate performance estimates, and hidden stratification, where errors on minority subpopulations are obscured by aggregate metrics. To address these issues, we propose a unified evaluation and training framework for spatially correlated data. We introduce Structure-Aware Stratified Partitioning (SASP), which constructs validation splits that reduce spatiotemporal leakage while preserving meaningful class balance, and Curriculum Distributionally Robust Optimization (CDRO), a curriculum-based relaxation of distributionally robust training that stabilizes optimization under these stricter splits. Across multiple benchmarks, this combination yields consistently improved generalization, more reliable confidence calibration, and exposes failure modes that remain hidden under conventional random-split evaluation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。