EAAI Journal 2025 Journal Article
Time Series Language Model for descriptive caption generation
- Mohamed Trabelsi
- Aidan Boyd
- Jin Cao
- Huseyin Uzunalioglu
The automatic generation of representative natural language descriptions for observable patterns in time series data enhances interpretability, simplifies analysis and increases the utility of temporal data across various engineering applications. In finance, for example, automated descriptions enable analysts to detect market trends and guide investment decisions. Similarly, in manufacturing, automated captioning helps detect anomalies in production systems, mitigating the risk of equipment failures and ensuring smooth operations. While pre-trained foundation models have made considerable progress in Natural Language Processing (NLP) and Computer Vision (CV), their application to time series analysis has been hindered by data scarcity. Although several Large Language Model (LLM)-based methods have been proposed for time series forecasting, time series captioning is under-explored in the context of LLMs. In this paper, we introduce a novel model, Time Series Language Model (TSLM), designed specifically for time series captioning. TSLM operates as an encoder–decoder model, leveraging both text prompts and time series data representations to capture subtle temporal patterns across multiple phases and generate precise textual descriptions of time series inputs. TSLM addresses the data scarcity problem in time series captioning by first leveraging an in-context prompting synthetic data generation, and second denoising the generated data via a novel cross-modal dense retrieval scoring applied to time series-caption pairs. Experimental findings on various time series captioning datasets demonstrate that TSLM outperforms existing state-of-the-art approaches by a significant margin.