NUM2EVENT: Interpretable Event Reasoning from Numerical time-series
This work addresses the challenge of making LLMs interpretably reason about events from numerical data, which is incremental as it builds on existing multimodal capabilities but introduces a new task and method.
The paper tackles the problem of inferring interpretable structured events from numerical time-series data, where existing methods focus on forecasting without explaining reasoning. It introduces a framework that outperforms LLM baselines in event-level precision and recall on multi-domain datasets.
Large language models (LLMs) have recently demonstrated impressive multimodal reasoning capabilities, yet their understanding of purely numerical time-series signals remains limited. Existing approaches mainly focus on forecasting or trend description, without uncovering the latent events that drive numerical changes or explaining the reasoning process behind them. In this work, we introduce the task of number-to-event reasoning and decoding, which aims to infer interpretable structured events from numerical inputs, even when current text is unavailable. To address the data scarcity and semantic alignment challenges, we propose a reasoning-aware framework that integrates an agent-guided event extractor (AGE), a marked multivariate Hawkes-based synthetic generator (EveDTS), and a two-stage fine-tuning pipeline combining a time-series encoder with a structured decoder. Our model explicitly reasons over numerical changes, generates intermediate explanations, and outputs structured event hypotheses. Experiments on multi-domain datasets show that our method substantially outperforms strong LLM baselines in event-level precision and recall. These results suggest a new direction for bridging quantitative reasoning and semantic understanding, enabling LLMs to explain and predict events directly from numerical dynamics.