激活很重要：面向视觉语言模型 OOD 检测的测试时激活负标签
Activation Matters: Test-time Activated Negative Labels for OOD Detection with Vision-Language Models
对做开放世界视觉识别的人，这篇给出了一个免训练、测试时在线挖掘高激活负标签的 OOD 检测流程：用测试样本在语料标签上的累积分配概率构造标签激活指标，并配合激活感知打分函数，可直接插到 CLIP 类 VLM 的零样本推理管线上，替代固定的远距离负标签集合。
原文依据 we propose Test-time Activated Negative Labels (TANL) by dynamically evaluating activation levels across the corpus dataset
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论分布外（OOD）检测旨在识别偏离分布内（ID）的样本。一类流行的流程通过引入远离 ID 类别的负标签，并依据样本到这些标签的距离进行 OOD 检测。然而，这类标签在 OOD 样本上可能激活较弱，无法刻画 OOD 特征。为此，我们提出测试时激活负标签（TANL），在测试过程中动态评估整个语料数据集上的激活水平，并挖掘具有高激活响应的候选标签。具体而言，TANL 在线识别高置信度的测试图像，并在语料上累积其分配概率，从而构造标签激活指标。该指标利用历史测试样本自适应地对齐测试分布，从而能够选出分布自适应的激活负标签。通过进一步挖掘当前测试批次内的激活信息，我们引入了一种更细粒度、批次自适应的变体。为充分利用标签激活知识，我们提出了一种激活感知的打分函数，强调激活更强的负标签，从而提升性能并增强其对标签数量的鲁棒性。我们的 TANL 无需训练、测试高效，并有理论依据支撑。在多种骨干网络和广泛任务设置上的实验验证了其有效性。值得注意的是，在大规模 ImageNet 基准上，TANL 将 FPR95 从 17.5% 显著降至 9.8%。代码将公开。
英文原文摘要 1625 字符
Out-of-distribution (OOD) detection aims to identify samples that deviate from in-distribution (ID). One popular pipeline addresses this by introducing negative labels distant from ID classes and detecting OOD based on their distance to these labels.However, such labels may present poor activation on OOD samples, failing to capture the OOD characteristics. To address this, we propose \underline T est-time \underline A ctivated \underline N egative \underline L abels (TANL) by dynamically evaluating activation levels across the corpus dataset and mining candidate labels with high activation responses in the testing process. Specifically, TANL identifies high-confidence test images online and accumulates their assignment probabilities over the corpus to construct a label activation metric.Such a metric leverages historical test samples to adaptively align with the test distribution, enabling the selection of distribution-adaptive activated negative labels. By further exploring the activation information within the current testing batch, we introduce a more fine-grained, batch-adaptive variant. To fully utilize label activation knowledge, we propose an activation-aware score function that emphasizes negative labels with stronger activations, boosting performance and enhancing its robustness to the label number.Our TANL is training-free, test-efficient, and grounded in theoretical justification. Experiments on diverse backbones and wide task settings validate its effectiveness.Notably, on the large-scale ImageNet benchmark, TANL significantly reduces the FPR95 from 17.5% to 9.8%.Codes will be released.