发表机构
University of York; Cyprus University of Technology（约克大学; 塞浦路斯理工大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对目标检测中的不确定性估计问题，提出轻量级后验证据元模型GRACE，冻结基础检测器，通过显著性引导课程学习估计定位不确定性，无需修改检测器，显著提升对抗鲁棒性。
AI 中文摘要
在可能发生分布/协变量偏移和对抗性攻击时，可靠的不确定性估计对于部署目标检测器至关重要。现有方法通常需要重新训练检测器、修改架构或进行重复推理，这可能不可行或产生显著开销。我们引入了一种轻量级的后验证据元模型，该模型在保持基础检测器冻结的同时，学习何时应将目标定位视为不确定。我们的方法自动识别与定位相关的特征，并使用显著性引导的修改来构建难度递增的课程。检测级目标结合定位误差、修改级别和预测不稳定性，引导证据元模型为每个预测边界框估计不确定性。我们的方法无需对检测器进行任何更改，并保留其原始定位输出。在对抗性攻击和评估强度下，GRACE在某些情况下相对于最强比较器将TP-FP AUROC提高了22%，同时保持了分布内检测性能。
英文摘要
Reliable uncertainty estimation is essential for deploying object detectors when distribution/covariate shift and adversarial attacks may occur. Existing approaches often require detector retraining, architectural modification, or repeated inference, which may be infeasible or incur significant overheads. We introduce a lightweight post-hoc evidential meta-model that learns when object localisations should be considered uncertain while keeping the base detector frozen. Our approach automatically identifies localisation-relevant features and uses saliency-guided modification to construct an increasingly challenging curriculum. Detection-level targets combine localisation error, modification level, and prediction instability to guide an evidential meta-model to estimate uncertainty for each predicted bounding box. Our approach requires no changes to the detector and preserves its original localisation outputs. Across adversarial attacks and evaluated strengths, GRACE improves TP-FP AUROC by 22% relative to the strongest comparator in some cases while maintaining in-distribution detection performance.
