Subject-exclusive cross-validation is the standard evaluation protocol for facial Action Unit (AU) detection, yet reported improvements are often small. We show that cross-validation itself introduces measurable stochastic variance. On BP4D+, repeated 3-fold subject-exclusive splits produce an empirical noise floor of $\pm 0.065$ in average F1, with substantially larger variation for low-prevalence AUs. Operating-point metrics such as F1 fluctuate more than threshold-independent measures such as AUC, and model ranking can change under different fold assignments. We further evaluate cross-dataset robustness using a Leave-One-Dataset-Out (LODO) protocol across five AU datasets. LODO removes partition randomness and exposes domain-level instability that is not visible under single-dataset cross-validation. Together, these results suggest that gains often reported in cross-fold validation may fall within protocol variance. Leave-one-dataset-out cross-validation yields more stable and interpretable findings

翻译：面向动作单元（AU）检测，基于被试互斥的交叉验证是标准评估协议，然而文献中报告的改进往往较小。本研究证明，交叉验证本身会引入可测量的随机方差。在BP4D+数据集上，重复进行3折被试互斥的分割策略观测到平均F1指标的实证噪声地板为$\pm 0.065$，其中低流行率AU的波动幅度更大。以操作点为基础的指标（如F1）比阈值无关指标（如AUC）表现出更显著的波动，且不同折分配下模型排序可能发生变化。我们进一步采用留一数据集（LODO）协议在五个AU数据集上评估跨数据集鲁棒性。LODO方法消除了分区随机性，揭示了单一数据集交叉验证无法观测的域级不稳定性。综合这些结果表明，交叉验证中常报道的性能提升可能处于协议方差范围内。留一数据集交叉验证可产生更稳定且可解释的结果。