Graph2TS: Structure-Controlled Time Series Generation via Quantile-Graph VAEs

TL;DR

Graph2TS introduces a novel approach for time series generation by separating global structure from local stochastic variations using quantile-graph VAEs, leading to more faithful and controllable synthetic signals.

Contribution

The paper presents a structure-residual perspective and a quantile-graph conditioned VAE for improved, structure-controlled time series generation.

Findings

Better distributional fidelity on diverse datasets

Enhanced temporal alignment and representativeness

Outperforms diffusion and GAN baselines

Abstract

Although recent generative models can produce time series with close marginal distributions, they often face a fundamental tension between preserving global temporal structure and modeling stochastic local variations, particularly for highly volatile signals with weak or irregular periodicity. Direct distribution matching in such settings can amplify noise or suppress meaningful temporal patterns. In this work, we propose a structure-residual perspective on time-series generation, viewing temporal data as the combination of a structural backbone and stochastic residual dynamics, thereby motivating the separation of global organization from sample-level variability. Based on this insight, we represent time-series structure using a quantile-based transition graph that compactly captures global distributional and temporal dependencies. Building on this representation, we propose Graph2TS, a quantile-graph conditioned variational autoencoder that performs cross-modal generation from structural graphs to time series. By conditioning generation on structure rather than labels or metadata, the model preserves global temporal organization while enabling controlled stochastic variation. Experiments on diverse datasets, including sunspot, electricity load, ECG, and EEG signals, demonstrate improved distributional fidelity, temporal alignment, and representativeness compared to diffusion- and GAN-based baselines, highlighting structure-controlled and cross-modal generation as a promising direction for time-series modeling.