发表机构
Cancer Research UK Cambridge Institute; University of Cambridge（英国癌症研究中心剑桥研究所; 剑桥大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出视觉绊线框架，利用模型行为的时间不稳定性（如表示漂移、预测振荡等）来预判深度视觉系统的故障，实验证明其比传统不确定性方法提供更早更准的警告。
AI 中文摘要
深度视觉系统尽管在基准测试中表现强劲，但仍然容易受到损坏、遮挡和分布偏移的影响。现有的可靠性方法通常在单个时间步评估不确定性，并未显式建模系统如何走向失败。我们引入了视觉绊线（Visual Tripwires），一种预测性可靠性框架，利用模型行为中的时间不稳定性来预判即将发生的故障。我们的核心假设是，预测退化通过潜在表示、预测轨迹和注意力结构的可测量变化逐步发展。视觉绊线通过表示漂移、预测振荡、轨迹曲率和注意力熵来捕捉这些变化。一个轻量级的绊线预测器在时间窗口内聚合这些信号，以估计在未来预测范围内发生故障的概率。跨多个数据集、架构和渐进扰动设置的实验表明，所提出的不稳定信号在预测退化之前出现，并提供比传统不确定性估计方法更早、更准确的故障警告。这些结果表明，时间不稳定性包含有关未来模型可靠性的有用信息，并为深度视觉系统中的早期预警提供了实用基础。
英文摘要
Deep vision systems remain vulnerable to corruption, occlusion, and distribution shift despite strong benchmark performance. Existing reliability methods typically evaluate uncertainty at individual time steps and do not explicitly model how a system progresses toward failure. We introduce Visual Tripwires, a predictive reliability framework that uses temporal instability in model behaviour to anticipate impending failure. Our central hypothesis is that predictive degradation develops progressively through measurable changes in latent representations, prediction trajectories, and attention structure. Visual Tripwires captures these changes using representation drift, prediction oscillation, trajectory curvature, and attention entropy. A lightweight tripwire predictor aggregates these signals over a temporal window to estimate the probability of failure within a future prediction horizon. Experiments across multiple datasets, architectures, and progressive perturbation settings show that the proposed instability signals emerge before predictive degradation and provide earlier and more accurate failure warnings than conventional uncertainty estimation methods. These results demonstrate that temporal instability contains useful information about future model reliability and provides a practical basis for early warning in deep vision systems.
