波的语法：通过神经符号VLM智能体实现可解释的多变量时间序列事件检测
Grammar of the Wave: Towards Explainable Multivariate Time Series Event Detection via Neuro-Symbolic VLM Agents
AI总结：
提出语言引导的时间序列事件检测（TSED）任务，通过事件逻辑树（ELT）将文本描述转化为结构化时序逻辑，并构建神经符号VLM智能体SELA，实现零/少样本事件检测与可解释推理。
摘要：
时间序列事件检测（TSED）旨在定位时间序列数据中具有语义意义的事件，在高风险领域具有关键应用。与统计异常不同，事件通常由自然语言描述定义，且跨多个物理通道具有内部时序逻辑结构。然而，在现实场景中，密集的事件标注成本高昂，使得纯监督学习困难。我们引入了语言引导的TSED，该设置中模型被赋予文本事件描述，并必须在几乎没有标注数据的情况下将其映射到多变量信号中的区间。为了解决这个问题，我们提出了事件逻辑树（ELT），一种知识表示框架，将语言描述转化为信号基元上的结构化时序逻辑。基于ELT，我们提出了SELA，一种神经符号VLM智能体框架，它从信号可视化中迭代地接地基元，并在ELT约束下组合它们，产生事件区间和忠实的树状结构解释。我们进一步发布了跨能源和气候领域的真实世界基准，包含专家知识和标注。实验表明，SELA优于监督微调和现有的零/少样本时间序列推理基线。
英文摘要：
Time Series Event Detection (TSED) aims to localize semantically meaningful events in time series data, with critical applications in high-stakes domains. Unlike statistical anomalies, events are often defined by natural-language descriptions with internal temporal-logic structures across multiple physical channels. However, in real-world settings, dense event annotations are expensive to obtain, making purely supervised learning difficult. We introduce Language-guided TSED, a setting where a model is given textual event descriptions and must ground them to intervals in multivariate signals with little or no labeled data. To address this problem, we propose Event Logic Tree (ELT), a knowledge representation framework that converts linguistic descriptions into structured temporal logic over signal primitives. Building on ELT, we present SELA, a neuro-symbolic VLM agent framework that iteratively grounds primitives from signal visualizations and composes them under ELT constraints, producing both event intervals and faithful tree-structured explanations. We further release a real-world benchmark across energy and climate domains with expert knowledge and annotations. Experiments show that SELA improves over supervised fine-tuning and existing zero/few-shot time series reasoning baselines.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
