BRIDGE: Bootstrapping Text to Control Time-Series Generation via Multi-Agent Iterative Optimization and Diffusion Modeling

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Li, Hao, Huang, Yu-Hao, Xu, Chang, Schlegel, Viktor, Jiang, Renhe, Batista-Navarro, Riza, Nenadic, Goran, Bian, Jiang
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914023585349632
author Li, Hao
Huang, Yu-Hao
Xu, Chang
Schlegel, Viktor
Jiang, Renhe
Batista-Navarro, Riza
Nenadic, Goran
Bian, Jiang
author_facet Li, Hao
Huang, Yu-Hao
Xu, Chang
Schlegel, Viktor
Jiang, Renhe
Batista-Navarro, Riza
Nenadic, Goran
Bian, Jiang
contents Time-series Generation (TSG) is a prominent research area with broad applications in simulations, data augmentation, and counterfactual analysis. While existing methods have shown promise in unconditional single-domain TSG, real-world applications demand for cross-domain approaches capable of controlled generation tailored to domain-specific constraints and instance-level requirements. In this paper, we argue that text can provide semantic insights, domain information and instance-specific temporal patterns, to guide and improve TSG. We introduce ``Text-Controlled TSG'', a task focused on generating realistic time series by incorporating textual descriptions. To address data scarcity in this setting, we propose a novel LLM-based Multi-Agent framework that synthesizes diverse, realistic text-to-TS datasets. Furthermore, we introduce BRIDGE, a hybrid text-controlled TSG framework that integrates semantic prototypes with text description for supporting domain-level guidance. This approach achieves state-of-the-art generation fidelity on 11 of 12 datasets, and improves controllability by up to 12% on MSE and 6% MAE compared to no text input generation, highlighting its potential for generating tailored time-series data.
format Preprint
id arxiv_https___arxiv_org_abs_2503_02445
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle BRIDGE: Bootstrapping Text to Control Time-Series Generation via Multi-Agent Iterative Optimization and Diffusion Modeling
Li, Hao
Huang, Yu-Hao
Xu, Chang
Schlegel, Viktor
Jiang, Renhe
Batista-Navarro, Riza
Nenadic, Goran
Bian, Jiang
Machine Learning
Computation and Language
Multiagent Systems
Time-series Generation (TSG) is a prominent research area with broad applications in simulations, data augmentation, and counterfactual analysis. While existing methods have shown promise in unconditional single-domain TSG, real-world applications demand for cross-domain approaches capable of controlled generation tailored to domain-specific constraints and instance-level requirements. In this paper, we argue that text can provide semantic insights, domain information and instance-specific temporal patterns, to guide and improve TSG. We introduce ``Text-Controlled TSG'', a task focused on generating realistic time series by incorporating textual descriptions. To address data scarcity in this setting, we propose a novel LLM-based Multi-Agent framework that synthesizes diverse, realistic text-to-TS datasets. Furthermore, we introduce BRIDGE, a hybrid text-controlled TSG framework that integrates semantic prototypes with text description for supporting domain-level guidance. This approach achieves state-of-the-art generation fidelity on 11 of 12 datasets, and improves controllability by up to 12% on MSE and 6% MAE compared to no text input generation, highlighting its potential for generating tailored time-series data.
title BRIDGE: Bootstrapping Text to Control Time-Series Generation via Multi-Agent Iterative Optimization and Diffusion Modeling
topic Machine Learning
Computation and Language
Multiagent Systems
url https://arxiv.org/abs/2503.02445