AI 中文总结
本文针对现有反事实解释不符合现实实例的问题，结合可解释AI与数据管理思路，利用拒绝约束生成现实反事实，经多数据集验证，该方案兼顾现实性与距离、多样性，且优化后搜索效率高。
AI 中文摘要
在可解释人工智能领域，分类结果常通过反事实（简称CFs）进行解释，反事实是对实例的（理想情况下较小的）扰动，能使分类标签发生改变，这类反事实可作为预测的解释，明确重要特征。现有可解释性解决方案通常旨在最小化反事实与原始实例的距离以保证针对性，和/或最大化反事实的多样性以覆盖预测背后的多个原因。本文指出，在追求这些目标时，现有最先进的可解释性解决方案可能且经常生成不符合现实实例的反事实解释，这限制了其在实践中的适用性和实用性。为解决该问题，本文将可解释人工智能的思路与数据管理的思路相结合，具体而言，通过针对示例数据集（如训练集）的逻辑约束来捕捉反事实的现实性，本文关注的这类约束属于关系数据库中广泛研究的拒绝约束类别。在算法层面，本文将可解释人工智能解决方案与适配到该特殊场景的数据清理思路相结合，将反事实转换为现实的反事实。在四个数据集上开展的大量实验验证，本文的解决方案在实现现实性的同时，仅在距离和多样性方面做出了相对较小的妥协；还验证了本文为加快反事实搜索而开发的专用优化方法确实非常有效。
英文摘要
In the realm of Explainable AI, classification results are often explained via counterfactuals (CFs for short), which are (ideally small) perturbations to an instance that lead to a change of classification label. Such CFs may serve as explanations for the prediction, pinpointing the features that were important. Existing explainability solutions typically aim at minimizing the distance of CFs from the original instance so that they are specific to it, and/or maximizing the diversity of CFs to cover multiple facets of the reasons underlying the prediction. In this paper, we note that in pursuing these aims, state-of-the-art explainability solutions may (and often do) yield counterfactual explanations that do not correspond to realistic instances. This limits their applicability and usefulness in practice. To remedy this, we combine ideas from Explainable AI with ideas from data management. Specifically, we capture realism of CFs via logical constraints that hold with respect to a dataset of examples (e.g., training set); the class of such constraints that we focus on is that of denial constraints, extensively studied in the context of relational databases. Algorithmically, we then combine explainable AI solutions to yield CFs, with ideas from data cleaning that we adapt to this unique setting, to transform CFs into realistic ones. Extensive experiments across four datasets validate that our solutions achieve realism with relatively minor compromise in terms of distance and diversity. They further validate that the dedicated optimizations that we have developed to speed up the search for CFs are indeed highly effective.
Journal refProc. 32nd ACM SIGKDD Conf. on Knowledge Discovery and Data Mining (KDD '26), Jeju Island, South Korea, Aug 2026, pp. 33-44
