# Forging Time Series with Language: A Large Language Model Approach to Synthetic Data Generation

arXiv: https://arxiv.org/abs/2505.17103 · HTML: https://mlsanity.com/paper/2505.17103/

## Key facts

- Cited by 7 papers in the map (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 21 of its 31 references that are in the map.
- First posted on arXiv 21 May 2025.

## Placed in

- Synthetic data and self-improvement for LLMs / [Model collapse from synthetic data](https://mlsanity.com/synth/model-collapse-from-synthetic-data/)

## Builds on (most cited first, 21 of 21)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecasting](https://mlsanity.com/paper/1912.09363/) (2019)
- [Chronos: Learning the Language of Time Series](https://mlsanity.com/paper/2403.07815/) (2024)
- [Time-LLM: Time Series Forecasting by Reprogramming Large Language Models](https://mlsanity.com/paper/2310.01728/) (2023)
- [One Fits All:Power General Time Series Analysis by Pretrained LM](https://mlsanity.com/paper/2302.11939/) (2023)
- [Large Language Models Are Zero-Shot Time Series Forecasters](https://mlsanity.com/paper/2310.07820/) (2023)
- [Language Models are Realistic Tabular Data Generators](https://mlsanity.com/paper/2210.06280/) (2022)
- [Foundation Models for Time Series Analysis: A Tutorial and Survey](https://mlsanity.com/paper/2403.14735/) (2024)
- [Neural SDEs as Infinite-Dimensional GANs](https://mlsanity.com/paper/2102.03657/) (2021)
- [TimeGPT-1](https://mlsanity.com/paper/2310.03589/) (2023)
- [TimeVAE: A Variational Auto-Encoder for Multivariate Time Series Generation](https://mlsanity.com/paper/2111.08095/) (2021)
- [Enabling Language Models to Fill in the Blanks](https://mlsanity.com/paper/2005.05339/) (2020)
- [Tabular Transformers for Modeling Multivariate Time Series](https://mlsanity.com/paper/2011.01843/) (2020)
- [Vector Quantized Time Series Generation with a Bidirectional Prior Model](https://mlsanity.com/paper/2303.04743/) (2023)
- [Deep Latent State Space Models for Time-Series Generation](https://mlsanity.com/paper/2212.12749/) (2022)
- [Conditional GAN for timeseries generation](https://mlsanity.com/paper/2006.16477/) (2020)
- [Towards Generating Real-World Time Series Data](https://mlsanity.com/paper/2111.08386/) (2021)
- [TSGBench: Time Series Generation Benchmark](https://mlsanity.com/paper/2309.03755/) (2023)
- [Generating multivariate time series with COmmon Source CoordInated GAN (COSCI-GAN)](https://mlsanity.com/paper/2205.13741/) (2022)
- [scikit-fda: A Python Package for Functional Data Analysis](https://mlsanity.com/paper/2211.02566/) (2022)
- [FAStEN: An Efficient Adaptive Method for Feature Selection and Estimation in High-Dimensional Functional Regressions](https://mlsanity.com/paper/2303.14801/) (2023)

## Built on by (most cited first, 7 of 7)

- [Time Series Reasoning via Process-Verifiable Thinking Data Synthesis and Scheduling for Tailored LLM Reasoning](https://mlsanity.com/paper/2602.07830/) (2026)
- [E4GEN: Event-level Explainable Extreme-Enhanced Time-series Generation](https://mlsanity.com/paper/2606.01634/) (2026)
- [SynEnergy: Anomaly Semantic-Guided Diffusion for Synthetic Energy Data Generation](https://mlsanity.com/paper/2608.03087/) (2026)
- [Redefining Maritime Anomaly Detection via Equation-Grounded Synthetic Anomalies](https://mlsanity.com/paper/2606.29721/) (2026)
- [Synthetic Worlds for Temporal Evaluation and Knowledge Updating in LLMs](https://mlsanity.com/paper/2609.00184/) (2026)
- [CodeTS: Verifiable Text-to-Time Series Generation via Executable Code](https://mlsanity.com/paper/2609.15393/) (2026)
- [Cross-View Time-Frequency Diffusion for Extreme-Aware Time-Series Generation](https://mlsanity.com/paper/2610.17407/) (2026)

## Abstract (arXiv)

SDForger is a flexible and efficient framework for generating high-quality multivariate time series using LLMs. Leveraging a compact data representation, SDForger provides synthetic time series generation from a few samples and low-computation fine-tuning of any autoregressive LLM. Specifically, the framework transforms univariate and multivariate signals into tabular embeddings, which are then encoded into text and used to fine-tune the LLM. At inference, new textual embeddings are sampled and decoded into synthetic time series that retain the original data's statistical properties and temporal dynamics. Across a diverse range of datasets, SDForger outperforms existing generative models in many scenarios, both in similarity-based evaluations and downstream forecasting tasks. By enabling textual conditioning in the generation process, SDForger paves the way for multimodal modeling and the streamlined integration of time series with textual information. The model is open-sourced at https://github.com/IBM/fms-dgt/tree/main/fms_dgt/public/databuilders/time_series.