Detection of false positives by human reviewers depends on the face recognition algorithm and its security configuration

查看原文 / View Original 编辑 / Edit

Computer Science > Computer Vision and Pattern Recognition arXiv:2610.33045 (cs) [Submitted on 16 Oct 2026] Title: Detection of false positives by human reviewers depends on the face recognition algorithm and its security configuration Authors: Esha Edakoth, John J. Howard, Isabelle M. Shuggi, Yevgeniy B. Sirotin, Robert Wood.

计算机科学 > 计算机视觉与模式识别 arXiv:2610.33045 (cs) [提交于 2026 年 10 月 16 日] 标题：人类审查员对误报的检测取决于人脸识别算法及其安全配置作者：Esha Edakoth, John J. Howard, Isabelle M. Shuggi, Yevgeniy B. Sirotin, Robert Wood。

Abstract: Integration of artificial intelligence (AI) technology into identity workflows, such as automated face recognition (FR), has enhanced operational capabilities, with FR algorithms achieving super-human scale and accuracy. However, these gains can make human oversight of FR more difficult, raising questions about the effectiveness of these human-AI teams.

摘要：将人工智能 (AI) 技术集成到身份工作流程中（例如自动人脸识别 (FR)）增强了操作能力，使 FR 算法达到了超越人类的规模和准确性。然而，这些进步可能会使人类对 FR 的监督变得更加困难，从而引发了关于这些“人机协作”团队有效性的质疑。

We measured human reviewers’ ability to detect false positive errors made by commercial FR algorithms across varying security configurations. Using a novel dataset, the Difficult Facial Image Comparison Test (DFICT), derived from false positives made by four comparable commercial face recognition algorithms, we measured the ability of 801 human reviewers to detect algorithmic false positives across three security configurations, representing operationally relevant biometric thresholds.

我们测量了人类审查员在不同安全配置下检测商业 FR 算法所产生的误报错误的能力。我们使用了一个名为“困难人脸图像比较测试”(DFICT) 的新颖数据集，该数据集源自四种可比的商业人脸识别算法产生的误报，并测量了 801 名人类审查员在代表操作相关生物识别阈值的三种安全配置下检测算法误报的能力。

Our findings show that human reviewers can reduce algorithmic false positives, but performance declines at stricter security thresholds. Further, reviewer performance varied by algorithm indicating that some algorithms are more compatible with human oversight. We suggest that biometric systems should be designed to account for human reviewers’ variable effectiveness, particularly in high-security applications.

我们的研究结果表明，人类审查员可以减少算法误报，但在更严格的安全阈值下，其表现会下降。此外，审查员的表现因算法而异，这表明某些算法与人工监督的兼容性更好。我们建议生物识别系统的设计应考虑到人类审查员效能的可变性，特别是在高安全性应用中。

Our findings should be considered during algorithm selection and inform human oversight strategies to minimize AI-related risks in operational FR deployments.

我们的研究结果应在算法选择过程中予以考虑，并为人工监督策略提供参考，以最大限度地减少操作性 FR 部署中与 AI 相关的风险。