它把解释当HP实际原因，用SCM建模输入依赖，搜索算法保证完备且最小，还发现忽视依赖会多出14.9%虚假原因。
该论文将神经网络预测的解释形式化为 Halpern-Pearl 实际原因，并用布尔结构因果模型（SCM）建模输入依赖。算法结合边界传播与分支定界，提供完备性和最小性保证。实验中，它在搜索空间达 2.3×10^13 个候选对、SCM 最多 28 个节点的实例上，于 180 秒内算出全部最小实际原因，规模上超过暴力搜索和 ILP 基线。案例研究显示，忽略输入依赖会使报告原因数量膨胀，其中 14.9% 在 SCM 下是虚假原因。
Computing Actual Causes for Neural Network Predictions under Structured Causal Inputs
Explaining the predictions of neural networks is a central challenge in trustworthy AI. Existing explanation methods, such as those based on feature attribution or minimal sufficient sets, typically treat input features as independent, which can yield misleading explanations when inputs exhibit structured dependencies. We address this by formalizing explanations as Halpern-Pearl (HP) actual causes, modeling input dependencies using Boolean Structural Causal Models (SCMs). We compute HP causes by applying bound propagation and branch-and-bound techniques, while providing formal guarantees of completeness and minimality. Our experiments show that we substantially outperform brute-force and ILP baselines in scalability, and outperform heuristic search as graph size grows, computing all minimal actual causes on instances with search spaces of up to $2.3\times10^{13}$ candidate (cause, contingency) pairs, on SCMs with up to 28 nodes, within a 180s per-instance budget. In a case study, we further show that ignoring input dependencies inflates the number of reported causes, 14.9% of which are spurious under our SCM.