mlsanityData to 14 Oct 2026

Vision-Enhanced Time Series Forecasting via Latent Diffusion Models

arXiv 2502.14887 ↗First posted 16 Feb 2025cs.CV cs.AIcited by 9

On the map: Diffusion models → Flow matching, time series and science

Abstract (arXiv)

Diffusion models have recently emerged as powerful frameworks for generating high-quality images. While recent studies have explored their application to time series forecasting, these approaches face significant challenges in cross-modal modeling and transforming visual information effectively to capture temporal patterns. In this paper, we propose LDM4TS, a novel framework that leverages the powerful image reconstruction capabilities of latent diffusion models for vision-enhanced time series forecasting. Instead of introducing external visual data, we are the first to use complementary transformation techniques to convert time series into multi-view visual representations, allowing the model to exploit the rich feature extraction capabilities of the pre-trained vision encoder. Subsequently, these representations are reconstructed using a latent diffusion model with a cross-modal conditioning mechanism as well as a fusion module. Experimental results demonstrate that LDM4TS outperforms various specialized forecasting models for time series forecasting tasks.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Placed in Flow matching, time series and science, one of 14 directions in Diffusion models. Builds on 50 of its 70 references that are in the map. First posted on arXiv 16 Feb 2025.

Builds on

50 of 70 references.

Closest ancestors · 10

Same direction, newest first

  1. 2024A Survey on Diffusion Models for Time Series and Spatio-Temporal Data29 Apr
  2. 2024Diffusion-TS: Interpretable Diffusion for General Time Series Generation4 Mar
  3. 2023AutoDiff: combining Auto-encoder and Diffusion model for tabular data synthesizing24 Oct
  4. 2023Mixed-Type Tabular Data Synthesis with Score-based Diffusion in Latent Space14 Oct
  5. 2023Predict, Refine, Synthesize: Self-Guiding Diffusion Models for Probabilistic Time Series Forecasting21 Jul
  6. 2023Non-autoregressive Conditional Diffusion Models for Time Series Prediction8 Jun
  7. 2023Diffusion Models for Time Series Applications: A Survey1 May
  8. 2023DiffSTG: Probabilistic Spatio-Temporal Graph Forecasting with Denoising Diffusion Models31 Jan
  9. 2021CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation7 Jul
  10. 2021Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecasting28 Jan

Other references · 40

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  4. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  5. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  6. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  7. 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,057
  8. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,169
  9. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 2,039
  10. 2016WaveNet: A Generative Model for Raw Audiocited by 1,722
  11. 2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,676
  12. 2020Reformer: The Efficient Transformercited by 1,668
28 more
  1. 2022Are Transformers Effective for Time Series Forecasting?cited by 1,424
  2. 2020DiffWave: A Versatile Diffusion Model for Audio Synthesiscited by 1,354
  3. 2021Palette: Image-to-Image Diffusion Modelscited by 1,300
  4. 2022FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecastingcited by 1,210
  5. 2022TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysiscited by 1,058
  6. 2017Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networkscited by 879
  7. 2020WaveGrad: Estimating Gradients for Waveform Generationcited by 622
  8. 2022Less Is More: Fast Multivariate Time Series Forecasting with Light Sampling-oriented MLP Structurescited by 160
  9. 2022ETSformer: Exponential Smoothing Transformers for Time-series Forecastingcited by 148
  10. 2023A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detectioncited by 145
  11. 2023UniTime: A Language-Empowered Unified Model for Cross-Domain Time Series Forecastingcited by 144
  12. 2024TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysiscited by 116
  13. 2024SparseTSF: Modeling Long-term Time Series Forecasting with 1k Parameterscited by 107
  14. 2023Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospectscited by 99
  15. 2023Generative Time Series Forecasting with Diffusion, Denoise, and Disentanglementcited by 90
  16. 2024VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasterscited by 88
  17. 2015Imaging Time-Series to Improve Classification and Imputationcited by 85
  18. 2023LightTS: Lightweight Time Series Classification with Adaptive Ensemble Distillation -- Extended Versioncited by 50
  19. 2023MSGNet: Learning Multi-Scale Inter-Series Correlations for Multivariate Time Series Forecastingcited by 50
  20. 2024Position: What Can Large Language Models Tell Us about Time Series Analysiscited by 47
  21. 2021ScoreGrad: Multivariate Probabilistic Time Series Forecasting with Continuous Energy-based Generative Modelscited by 43
  22. 2024Time-FFM: Towards LM-Empowered Federated Foundation Model for Time Series Forecastingcited by 26
  23. 2023SaDI: A Self-adaptive Decomposed Interpretable Framework for Electric Load Forecasting under Extreme Eventscited by 21
  24. 2023EHRDiff: Exploring Realistic EHR Synthesis with Diffusion Modelscited by 19
  25. 2024Attractor Memory for Long-Term Time Series Forecasting: A Chaos Perspectivecited by 17
  26. 2024ViTime: Foundation Model for Time Series Forecasting Powered by Vision Intelligencecited by 11
  27. 2024ST-LoRA: Low-rank Adaptation for Spatio-Temporal Forecastingcited by 9
  28. 2024Cross Space and Time: A Spatio-Temporal Unitized Model for Traffic Flow Forecastingcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 6.

9 papers.

  1. 2025Diffusion Models for Time Series Forecasting: A Surveycited by 11
  2. 2025From Samples to Scenarios: A New Paradigm for Probabilistic Forecastingcited by 10
  3. 2025Multimodal Conditioned Diffusive Time Series Forecastingcited by 9
  4. 2026Visual Reasoning over Time Series via Multi-Agent Systemcited by 2
  5. 2026VETime: Vision Enhanced Zero-Shot Time Series Anomaly Detectioncited by 0
  6. 2026From Coarse to Fine: Managing Temporal Granularity in Spatio-Temporal Data for Fine-Grained Traffic Predictioncited by 0
  7. 2026Beyond Uniform Tokens: Adaptive Compression for Time Series Language Modelscited by 0
  8. 2026Distribution-Aware Diffusion-LLM for Robust Ultra-Long-Term Time Series Forecastingcited by 0
  9. 2026ProtoFlow: Prototype-Guided Flow Matching for Multivariate Time Series Forecastingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026TriTS: Time Series Forecasting from a Multimodal Perspective
  2. 2026DiTS: Multimodal Diffusion Transformers Are Time Series Forecasters
  3. 2024Utilizing Image Transforms and Diffusion Models for Generative Modeling of Short and Long Time Series
  4. 2025Time-VLM: Exploring Multimodal Vision-Language Models for Augmented Time Series Forecasting
  5. 2025OccamVTS: Distilling Vision Models to 1% Parameters for Time Series Forecasting
  6. 2026MM-ISTS: Cooperating Irregularly Sampled Time Series Forecasting with Multimodal Vision-Text LLMs
  7. 2026Image Synthesis as an Intermediate for Controllable Time Series Generation
  8. 2026Enhancing few-shot time series forecasting with LLM-guided diffusion