AI 中文总结
针对时间序列异常检测中上下文学习忽略参考的问题，提出反事实监督训练方法Anlu，通过冻结基础模型加参考记忆和门控适配器，在TSB-AD-U上提升VUS-PR至0.607。
AI 中文摘要
时间序列模式是否异常通常取决于被监测过程的运行状态。一个缺失事件在一个状态下可能表示故障，而在另一个状态下可能是常规情况，仅凭查询本身可能无法确定适用于哪种状态。我们通过参考条件检测来研究时间序列异常检测（TSAD）中的上下文学习（ICL），其中参考记录提供了关于预期行为的证据，且模型参数在推理时保持不变。仅提供参考是不够的：当训练异常仅从查询中即可识别时，检测器可能会在忽略参考的情况下拟合其目标。因此，我们引入了反事实监督，它将一个查询与两个支持不同正常规则的参考配对，并在每个参考下对查询进行标注。在两个标注不一致的位置，任何忽略参考的检测器都无法同时拟合两个目标。Anlu通过向冻结的时间序列基础模型（TSFM）添加参考记忆和零初始化的门控适配器来学习这种监督，该模型预训练用于异常检测。在350个TSB-AD-U评估序列上，Anlu将冻结TSFM的平均VUS-PR从0.542提高到0.607。将参考替换为零会使Anlu的得分降至0.499。
英文摘要
Whether a time-series pattern is anomalous often depends on the operating regime of the monitored process. A missing event can signal a fault in one regime and be routine in another, and the query alone may not reveal which regime applies. We study in-context learning (ICL) for time series anomaly detection (TSAD) through reference-conditioned detection, where a reference record provides evidence about expected behavior and model parameters remain fixed at inference. Supplying the reference is not enough: when training anomalies are recognizable from the query alone, the detector can fit its targets while ignoring the reference. We therefore introduce counterfactual supervision, which pairs one query with two references that support different normal rules and labels the query under each. At positions where the two labels disagree, no detector that ignores the reference can fit both targets. Anlu learns from this supervision by adding a reference memory and zero-initialized gated adapters to a frozen time-series foundation model (TSFM) pretrained for anomaly detection. On the 350 TSB-AD-U evaluation sequences, Anlu raises the mean VUS-PR of the frozen TSFM from 0.542 to 0.607. Replacing the reference with zeros lowers Anlu's score to 0.499.
