Grammar of the Wave: Towards Explainable Multivariate Time Series Event Detection via Neuro-Symbolic VLM Agents
On the map: LLM agents and tool use → Financial trading and forecasting agents
Abstract (arXiv)
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.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Financial trading and forecasting agents, one of 16 directions in LLM agents and tool use. Builds on 16 of its 76 references that are in the map. First posted on arXiv 12 Mar 2026.
Builds on
16 of 76 references.
References · 16
- 2017Attention Is All You Needcited by 60,361
- 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,733
- 2017Axiomatic Attribution for Deep Networkscited by 3,384
- 2024Chronos: Learning the Language of Time Seriescited by 759
- 2023Large Language Models Are Zero-Shot Time Series Forecasterscited by 500
- 2024MOMENT: A Family of Open Time-series Foundation Modelscited by 482
- 2024Scaling Large Language Model-based Multi-Agent Collaborationcited by 241
- 2024Timer: Generative Pre-trained Transformers Are Large Time Series Modelscited by 208
- 2017DeepSleepNet: a Model for Automatic Sleep Stage Scoring based on Raw Single-Channel EEGcited by 157
- 2022Deep Learning for Time Series Anomaly Detection: A Surveycited by 131
- 2024ChatTS: Aligning Time Series with LLMs via Synthetic Data for Enhanced Understanding and Reasoningcited by 103
- 2019U-Time: A Fully Convolutional Network for Time Series Segmentation Applied to Sleep Stagingcited by 88
4 more
- 2023Deep Learning for Time Series Classification and Extrinsic Regression: A Current Surveycited by 68
- 2018Precision and Recall for Time Seriescited by 67
- 2024A Picture is Worth A Thousand Numbers: Enabling LLMs Reason about Time Series via Visualizationcited by 33
- 2020ABBA: Adaptive Brownian bridge-based symbolic aggregation of time seriescited by 13
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025NUM2EVENT: Interpretable Event Reasoning from Numerical time-series
- 2025AXIS: Explainable Time Series Anomaly Detection with Large Language Models
- 2025Towards Interpretable Time Series Foundation Models
- 2025Training-Free Time Series Classification via In-Context Reasoning with LLM Agents
- 2026Thoth: Mid-Training Bridges LLMs to Time Series Understanding
- 2024Towards Time Series Reasoning with LLMs
- 2025TimeCAP: Learning to Contextualize, Augment, and Predict Time Series Events with Large Language Model Agents
- 2025TS-Agent: Understanding and Reasoning Over Raw Time Series via Iterative Insight Gathering