Does Synthetic Data Help? Empirical Evidence from Deep Learning Time Series Forecasters
On the map: Synthetic data and self-improvement for LLMs → Model collapse from synthetic data
Abstract (arXiv)
Synthetic data has transformed language model training, yet its role in time series forecasting remains poorly understood. We present a large-scale empirical study: nine experiment groups, 4,218 runs systematically evaluating synthetic time series augmentation across five architectures, four synthetic signals and seven datasets. The effect is sharply architecture-conditional: channel-mixing models (TimesNet, iTransformer) benefit in the majority of trials, while channel-independent models (DLinear, PatchTST) are consistently degraded. In selected low-resource settings the gains are striking: TimesNet trained on only 10% of Weather data with synthetic augmentation surpasses the full-data baseline (4 of 16 sparsity-dataset combinations). Averaged across all architectures, augmentation hurts in 67% of trials. We further find that only the Seasonal-Trend generator reliably helps across the tested benchmarks, and that hard curriculum switching is actively harmful (+24% MSE degradation). These results provide concrete, actionable guidelines on how to use synthetic data: use synthetic augmentation with channel-mixing architectures, use gradual annealing schedules, and treat low-resource augmentation as architecture- and dataset-dependent. Code is available at hrefhttps://github.com/hugoiscracked/synthetic-ts/tree/main
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Model collapse from synthetic data, one of 13 directions in Synthetic data and self-improvement for LLMs. Builds on 15 of its 26 references that are in the map. First posted on arXiv 7 May 2026.
Builds on
15 of 26 references.
References · 15
- 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,169
- 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 2,039
- 2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,676
- 2022Are Transformers Effective for Time Series Forecasting?cited by 1,424
- 2022TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysiscited by 1,058
- 2023iTransformer: Inverted Transformers Are Effective for Time Series Forecastingcited by 931
- 2023A decoder-only foundation model for time-series forecastingcited by 692
- 2024Unified Training of Universal Time Series Forecasting Transformerscited by 614
- 2025Chronos-2: From Univariate to Universal Forecastingcited by 194
- 2021Unity Perception: Generate Synthetic Data for Computer Visioncited by 43
- 2024DataDreamer: A Tool for Synthetic Data Generation and Reproducible LLM Workflowscited by 43
- 2018Data augmentation using synthetic data for time series classification with deep residual networkscited by 38
3 more
- 2025Demystifying Synthetic Data in LLM Pre-training: A Systematic Study of Scaling Laws, Benefits, and Pitfallscited by 26
- 2024Towards a Theoretical Understanding of Synthetic Data in LLM Post-Training: A Reverse-Bottleneck Perspectivecited by 15
- 2019Using GANs for Sharing Networked Time Series Data: Challenges, Initial Promise, and Open Questionscited by 12
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026Mix, Don't Pick: Why Synthetic Corpus Composition Matters for Time Series Foundation Model Pretraining
- 2025Empowering Time Series Analysis with Synthetic Data: A Survey and Outlook in the Era of Foundation Models
- 2026Improving Machine Learning Performance with Synthetic Augmentation
- 2025TIMED: Adversarial and Autoregressive Refinement of Diffusion-Based Time Series Generation
- 2026Rethinking the Role of LLMs in Time Series Forecasting
- 2023Large Language Models Are Zero-Shot Time Series Forecasters
- 2024Chronos: Learning the Language of Time Series
- 2025Boosting Statistic Learning with Synthetic Data from Pretrained Large Models