通过平衡实验设计，精准解析模型预测背后的主效应与交互作用。
CUBE: Contrastive Understanding by Balanced Experiments
- 设计高低对比查询，系统化构造可解释的实验条件。
- 在合成与真实表格数据上准确还原模型学习到的关键影响结构。
- 适合需要高效筛选与迭代优化解释结果的研究者使用。
后置解释依赖于模型查询的组织方式。我们提出 CUBE，一种基于设计的框架，通过平衡的低-高探针解释训练好的预测模型。选定变量定义因素，设计特征级组合定义查询条件，模型预测被总结为因子对比。CUBE 报告主效应和成对交互作用，作为在声明的设计空间内平均与条件响应变化的受控读数。在合成与真实表格任务上的实验表明，CUBE 能恢复主导的学习效应结构，阐明查询效率下的可辨识性，并支持筛选-后续优化流程。
原文摘要 · Abstract (English)
Post-hoc explanation depends on how model queries are organized. We propose CUBE, a design-based framework that explains a trained predictive model through balanced low--high probes. Selected variables define factors, designed feature-level combinations define query conditions, and model predictions are summarized as factorial contrasts. CUBE reports main effects and pairwise interactions as controlled readings of average and conditional response changes over a declared design space. Experiments on synthetic and real tabular tasks show that CUBE recovers dominant learned effect structure, clarifies query-efficient identifiability, and supports screening--follow-up refinement.
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。