The ability to properly benchmark model performance in the face of spurious correlations is important to both build better predictors and increase confidence that models are operating as intended. We demonstrate that characterizing (as opposed to simply quantifying) model mistakes across subgroups is pivotal to properly reflect model biases, which are ignored by standard metrics such as worst-group accuracy or accuracy gap. Inspired by the hypothesis testing framework, we introduce SkewSize, a principled and flexible metric that captures bias from mistakes in a model's predictions. It can be used in multi-class settings or generalised to the open vocabulary setting of generative models. SkewSize is an aggregation of the effect size of the interaction between two categorical variables: the spurious variable representing the bias attribute and the model's prediction. We demonstrate the utility of SkewSize in multiple settings including: standard vision models trained on synthetic data, vision models trained on ImageNet, and large scale vision-and-language models from the BLIP-2 family. In each case, the proposed SkewSize is able to highlight biases not captured by other metrics, while also providing insights on the impact of recently proposed techniques, such as instruction tuning.

翻译：在存在伪相关性的情况下，正确评估模型性能对于构建更优的预测器及增强模型按预期运行的置信度至关重要。我们证明，刻画（而非简单量化）模型在各子组中的错误特征，对于准确反映模型偏差具有关键作用，而最差组准确率或准确率差距等标准指标往往忽略此类偏差。受假设检验框架启发，我们提出SkewSize——一种原则性且灵活的度量指标，能够从模型预测错误中捕捉偏差。该指标适用于多分类场景，并可推广至生成模型的开放词汇表设定。SkewSize本质上是两个分类变量交互效应大小的聚合：代表偏差属性的伪变量与模型预测变量。我们在多种场景中验证了SkewSize的实用性，包括：基于合成数据训练的标准视觉模型、基于ImageNet训练的视觉模型，以及BLIP-2系列的大规模视觉-语言模型。在所有案例中，所提出的SkewSize均能揭示其他指标未能捕捉的偏差，同时为近期提出的技术（如指令微调）的影响提供新的见解。