AI 中文总结
本文提出收敛时的扰动敏感性信号，无需群体标注与早停轮次即可识别虚假相关样本，用其重新平衡训练可将Waterbirds数据集最差群体准确率从57.3%提至80.8%。
AI 中文摘要
在包含虚假相关性的数据上通过经验风险最小化训练的模型，虽能达到较高的平均准确率，却在相关性不成立的子群体上失效。现有无需群体标注即可识别受影响样本的方法依赖训练早期的信号，这需要确定干预的轮次，该超参数通常用带群体标签的验证数据选择。本文表明，收敛后存在可用信号，此时损失无法区分两个群体：符合虚假相关性的样本由共享规则分类，其余样本通过针对单个输入的配置拟合，因此更脆弱。对收敛后模型的输入施加固定扰动，翻转后者预测的频率远高于前者。该方法每个训练样本仅需两次前向传播，全程无需群体标注，也无需早停轮次。用检测到的样本重新平衡训练，可将Waterbirds数据集上的最差群体准确率从57.3%提升至80.8%，而使用真实群体标签时该值为85.8%。
英文摘要
Models trained by empirical risk minimization on data containing spurious correlations achieve high average accuracy while failing on subpopulations where the correlation does not hold. Existing methods for identifying the affected samples without group annotations rely on signals from early training, which requires locating the epoch at which to intervene, a hyperparameter typically selected using group-labeled validation data. We show that a usable signal is available after convergence, when loss no longer distinguishes the two populations. Samples consistent with the spurious correlation are classified by a shared rule, while the remaining samples are fit through configurations specific to individual inputs and are correspondingly more fragile. Applying a fixed perturbation to a converged model's inputs flips the predictions of the latter far more often than the former. The resulting procedure requires two forward passes per training sample, no group annotations at any stage, and no early-stopping epoch. Using the detected samples to rebalance training raises worst-group accuracy on Waterbirds from 57.3% to 80.8%, against 85.8% with ground-truth group labels.
