Do Generative Models Keep Time? A Time-Aware Evaluation of Synthetic Sequential Tabular Data

• URL: http://arxiv.org/abs/2607.15606v1
• Date: Fri, 17 Jul 2026 03:59:35 GMT
• Title: Do Generative Models Keep Time? A Time-Aware Evaluation of Synthetic Sequential Tabular Data
• Abstract summary: We present a taxonomy-guided evaluation protocol for temporal fidelity.<n>We apply the protocol to eight generative models across thirteen datasets spanning six domains.
• Score: 28.948173366251286
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Synthetic sequential tabular data are increasingly used for privacy-preserving data sharing, yet a generator can reproduce every marginal and every foreign-key relationship while emitting timestamps that run backwards or repeat, and while sending entities along paths that no real entity followed. Conventional tabular evaluation, which pools records into static distributions, is blind to such failures. We present a taxonomy-guided evaluation protocol for temporal fidelity, in which the applicable measurements are determined by the data rather than fixed in advance. Each dataset is first characterized along four properties: how time is represented, whether observations are regularly sampled, whether trajectories are mutually dependent, and how the schema links entities to their histories. These properties determine which evaluation dimensions are meaningful. The protocol then measures timestamp validity, cross-sectional structure at aligned time points, within-entity dynamics, and time-varying relational structure, and recasts utility and privacy evaluation over trajectories rather than isolated rows. We apply the protocol to eight generative models across thirteen datasets spanning six domains. Rankings under conventional evaluation disagree substantially with those obtained under temporal evaluation, and the resulting failures are architecture-coherent rather than random. Temporal fidelity must therefore be measured on the time axis itself, rather than inferred from pooled record distributions.

Related papers

• LiFTER: A Grounded Neuro-Symbolic Microscope for Continuous-Time Dynamic Graph Forecasting [0.0]
  Continuous-time dynamic graph models predict future links by compressing past interactions into neural states.<n>LiFTER is a neuro-symbolic predictor that preserves observed interactions as grounded temporal facts.<n>LiFTER achieves competitive historical-negative forecasting and the highest macro explanation ac- curacy and deletion fidelity.
  arXiv  Detail & Related papers  (2026-08-07T03:39:09Z)
• DoTime: A Synthetic Benchmark Generator for Interventional and Counterfactual Time Series [3.2517081185249075]
  We introduce textbfDoTime, an open, scalable, and theoretically grounded generator of temporal structural causal models with interventions.<n>It adds capabilities absent from prior generators: continuous-time intervention emphwindows, counterfactual sampling modes with a positivity guard, regime-switching SCMs.<n>It demonstrates the suitability of the generator as a prior for a causal foundation model reference implementation.
  arXiv  Detail & Related papers  (2026-07-29T07:46:44Z)
• TimeLAVA: Learning-Agnostic Data Valuation for Time Series [48.76711740818533]
  TimeLAVA is a learning-agnostic framework that values temporal segments by their marginal contribution to minimize distributional discrepancy between evaluated and reference data.<n>We provide theoretical guarantees linking valuation to model-agnostic generalization and prove bounded sensitivity to contamination.
  arXiv  Detail & Related papers  (2026-06-17T06:16:16Z)
• Estimating Mutual Information between Time Series and Temporal Event Sequences Across Diverse Analysis Tasks [21.6590064392125]
  We propose a nonparametric mutual information estimator that measures the dependence between time series and event sequences.<n>Our method models the continuous-discrete duality of real-world time series to handle quantization and repeated-value artifacts.<n>Experiments on synthetic and real-world datasets show consistent improvements over existing methods in accuracy, robustness, and interpretability.
  arXiv  Detail & Related papers  (2026-06-01T02:56:50Z)
• It's TIME: Towards the Next Generation of Time Series Forecasting Benchmarks [87.7937890373758]
  Time series foundation models (TSFMs) are revolutionizing the forecasting landscape from specific dataset modeling to generalizable task evaluation.<n>We introduce TIME, a next-generation task-centric benchmark comprising 50 fresh datasets and 98 forecasting tasks.<n>We propose a novel pattern-level evaluation perspective that moves beyond traditional dataset-level evaluations based on static meta labels.
  arXiv  Detail & Related papers  (2026-02-12T16:31:01Z)
• Relational Conformal Prediction for Correlated Time Series [56.59852921638328]
  We address the problem of uncertainty quantification in time series by exploiting correlated sequences.<n>We propose a novel distribution-free approach based on conformal prediction framework and quantile regression.<n>Our approach provides accurate coverage and achieves state-of-the-art uncertainty quantification in relevant benchmarks.
  arXiv  Detail & Related papers  (2025-02-13T16:12:17Z)
• Evaluating Inter-Column Logical Relationships in Synthetic Tabular Data Generation [49.898152180805454]
  This paper proposes three evaluation metrics designed to assess the preservation of logical relationships.<n>We validate these metrics by assessing the performance of both classical and state-of-the-art generation methods on a real-world industrial dataset.
  arXiv  Detail & Related papers  (2025-02-06T13:13:26Z)
• From Link Prediction to Forecasting: Addressing Challenges in Batch-based Temporal Graph Learning [0.716879432974126]
  We show that the suitability of common batch-oriented evaluation depends on the datasets' characteristics.<n>For continuous-time temporal graphs, fixed-size batches create time windows with different durations, resulting in an inconsistent dynamic link prediction task.<n>For discrete-time temporal graphs, the sequence of batches can additionally introduce temporal dependencies that are not present in the data.
  arXiv  Detail & Related papers  (2024-06-07T12:45:12Z)
• Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling [24.368893944128086]
  Stemporal graph neural networks achieve striking results by representing relationships across time series as a graph. Most existing methods rely on the often unrealistic assumption that inputs are always available and fail to capture hidden dynamics when part of the data is missing.
  arXiv  Detail & Related papers  (2024-02-16T12:33:31Z)
• TimeSiam: A Pre-Training Framework for Siamese Time-Series Modeling [67.02157180089573]
  Time series pre-training has recently garnered wide attention for its potential to reduce labeling expenses and benefit various downstream tasks. This paper proposes TimeSiam as a simple but effective self-supervised pre-training framework for Time series based on Siamese networks.
  arXiv  Detail & Related papers  (2024-02-04T13:10:51Z)
• Time-series Generation by Contrastive Imitation [87.51882102248395]
  We study a generative framework that seeks to combine the strengths of both: Motivated by a moment-matching objective to mitigate compounding error, we optimize a local (but forward-looking) transition policy. At inference, the learned policy serves as the generator for iterative sampling, and the learned energy serves as a trajectory-level measure for evaluating sample quality.
  arXiv  Detail & Related papers  (2023-11-02T16:45:25Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.