AdaStop：用于 DNN 测试选择的成本感知提前停止

arXiv:2607.05461v1 Announce Type: new Abstract: Existing methods for testing deep neural networks (DNNs) primarily prioritize test inputs likely to reveal model faults under a fixed labeling budget.在实践中，选择预算很困难：测试太少会错过失败，而太多则会产生不必要的标签成本。这项工作研究了 DNN 测试中的停止问题。我们将测试制定为成本效益决策过程，其中标记输入会产生成本 $c$，发现故障会产生价值 $v$。 Based on this formulation, we introduce \textit{AdaStop}, a framework that estimates the marginal fault discovery rate during testing and stops labeling when the estimated rate falls below the threshold $\tau = c/v$. Experiments across multiple datasets, architectures, and selection strategies show that $65$--$84\%$ of faults can be discovered using only $9$--$31\%$ of the labeling budget.