VerbalTS: Generating Time Series from Texts

Shuqi Gu, Chuyue Li, Baoyu Jing, Kan Ren
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:20448-20476, 2025.

Abstract

Time series synthesis has become a foundational task in modern society, underpinning decision-making across various scenes. Recent approaches primarily generate time series from structured conditions, such as attribute-based metadata. However, these methods struggle to capture the full complexity of time series, as the predefined structures often fail to reflect intricate temporal dynamics or other nuanced characteristics. Moreover, constructing structured metadata requires expert knowledge, making large-scale data labeling costly and impractical. In this paper, we introduce VerbalTS, a novel framework for generating time series from unstructured textual descriptions, offering a more expressive and flexible solution to time series synthesis. To bridge the gap between unstructured text and time series data, VerbalTS employs a multi-focal alignment and generation framework, effectively modeling their complex relationships. Experiments on two synthetic and four real-world datasets demonstrate that VerbalTS outperforms existing methods in both generation quality and semantic alignment with textual conditions.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-gu25a, title = {{V}erbal{TS}: Generating Time Series from Texts}, author = {Gu, Shuqi and Li, Chuyue and Jing, Baoyu and Ren, Kan}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {20448--20476}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/gu25a/gu25a.pdf}, url = {https://proceedings.mlr.press/v267/gu25a.html}, abstract = {Time series synthesis has become a foundational task in modern society, underpinning decision-making across various scenes. Recent approaches primarily generate time series from structured conditions, such as attribute-based metadata. However, these methods struggle to capture the full complexity of time series, as the predefined structures often fail to reflect intricate temporal dynamics or other nuanced characteristics. Moreover, constructing structured metadata requires expert knowledge, making large-scale data labeling costly and impractical. In this paper, we introduce VerbalTS, a novel framework for generating time series from unstructured textual descriptions, offering a more expressive and flexible solution to time series synthesis. To bridge the gap between unstructured text and time series data, VerbalTS employs a multi-focal alignment and generation framework, effectively modeling their complex relationships. Experiments on two synthetic and four real-world datasets demonstrate that VerbalTS outperforms existing methods in both generation quality and semantic alignment with textual conditions.} }
Endnote
%0 Conference Paper %T VerbalTS: Generating Time Series from Texts %A Shuqi Gu %A Chuyue Li %A Baoyu Jing %A Kan Ren %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-gu25a %I PMLR %P 20448--20476 %U https://proceedings.mlr.press/v267/gu25a.html %V 267 %X Time series synthesis has become a foundational task in modern society, underpinning decision-making across various scenes. Recent approaches primarily generate time series from structured conditions, such as attribute-based metadata. However, these methods struggle to capture the full complexity of time series, as the predefined structures often fail to reflect intricate temporal dynamics or other nuanced characteristics. Moreover, constructing structured metadata requires expert knowledge, making large-scale data labeling costly and impractical. In this paper, we introduce VerbalTS, a novel framework for generating time series from unstructured textual descriptions, offering a more expressive and flexible solution to time series synthesis. To bridge the gap between unstructured text and time series data, VerbalTS employs a multi-focal alignment and generation framework, effectively modeling their complex relationships. Experiments on two synthetic and four real-world datasets demonstrate that VerbalTS outperforms existing methods in both generation quality and semantic alignment with textual conditions.
APA
Gu, S., Li, C., Jing, B. & Ren, K.. (2025). VerbalTS: Generating Time Series from Texts. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:20448-20476 Available from https://proceedings.mlr.press/v267/gu25a.html.

Related Material