mlsanityData to 7 Oct 2026

Multimodal Conditioned Diffusive Time Series Forecasting

arXiv 2504.19669 ↗First posted 28 Apr 2025cs.CLcited by 8

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

Abstract (arXiv)

Diffusion models achieve remarkable success in processing images and text, and have been extended to special domains such as time series forecasting (TSF). Existing diffusion-based approaches for TSF primarily focus on modeling single-modality numerical sequences, overlooking the rich multimodal information in time series data. To effectively leverage such information for prediction, we propose a multimodal conditioned diffusion model for TSF, namely, MCD-TSF, to jointly utilize timestamps and texts as extra guidance for time series modeling, especially for forecasting. Specifically, Timestamps are combined with time series to establish temporal and semantic correlations among different data points when aggregating information along the temporal dimension. Texts serve as supplementary descriptions of time series' history, and adaptively aligned with data points as well as dynamically controlled in a classifier-free manner. Extensive experiments on real-world benchmark datasets across eight domains demonstrate that the proposed MCD-TSF model achieves state-of-the-art performance.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Placed in Flow matching, time series and science, one of 14 directions in Diffusion models. Builds on 49 of its 69 references that are in the map. First posted on arXiv 28 Apr 2025.

Builds on

49 of 69 references.

Closest ancestors · 6

Same direction, newest first

  1. 2025Vision-Enhanced Time Series Forecasting via Latent Diffusion Models16 Feb
  2. 2024Retrieval-Augmented Diffusion Models for Time Series Forecasting24 Oct
  3. 2024MG-TSD: Multi-Granularity Time Series Diffusion Models with Guided Learning Process9 Mar
  4. 2024Diffusion-TS: Interpretable Diffusion for General Time Series Generation4 Mar
  5. 2021CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation7 Jul
  6. 2021Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecasting28 Jan

Other references · 43

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2023GPT-4 Technical Reportcited by 18,132
  4. 2024The Llama 3 Herd of Modelscited by 14,192
  5. 2020Denoising Diffusion Implicit Modelscited by 9,655
  6. 2022Classifier-Free Diffusion Guidancecited by 5,717
  7. 2024Qwen2.5 Technical Reportcited by 4,713
  8. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,141
  9. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 1,981
  10. 2020Reformer: The Efficient Transformercited by 1,658
  11. 2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,654
  12. 2022Are Transformers Effective for Time Series Forecasting?cited by 1,400
31 more
  1. 2022FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecastingcited by 1,200
  2. 2023Time-LLM: Time Series Forecasting by Reprogramming Large Language Modelscited by 680
  3. 2023One Fits All:Power General Time Series Analysis by Pretrained LMcited by 615
  4. 2023Large Language Models Are Zero-Shot Time Series Forecasterscited by 491
  5. 2022PromptCast: A New Prompt-based Learning Paradigm for Time Series Forecastingcited by 170
  6. 2023TEST: Text Prototype Aligned Embedding to Activate LLM's Ability for Time Seriescited by 165
  7. 2023UniTime: A Language-Empowered Unified Model for Cross-Domain Time Series Forecastingcited by 139
  8. 2017Conditional Time Series Forecasting with Convolutional Neural Networkscited by 121
  9. 2024TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysiscited by 113
  10. 2024ChatTime: A Unified Multimodal Time Series Foundation Model Bridging Numerical and Textual Datacited by 92
  11. 2023Generative Time Series Forecasting with Diffusion, Denoise, and Disentanglementcited by 89
  12. 2023Temporal Data Meets LLM -- Explainable Financial Time Series Forecastingcited by 77
  13. 2024Time-MMD: Multi-Domain Multimodal Dataset for Time Series Analysiscited by 65
  14. 2024AutoTimes: Autoregressive Time Series Forecasters via Large Language Modelscited by 63
  15. 2024TimeMachine: A Time Series is Worth 4 Mambas for Long-term Forecastingcited by 63
  16. 2024Time Series Forecasting with LLMs: Understanding and Enhancing Model Capabilitiescited by 48
  17. 2023ChiMed-GPT: A Chinese Medical Large Language Model with Full Training Regime and Better Alignment to Human Preferencescited by 43
  18. 2024Can LLMs Understand Time Series Anomalies?cited by 41
  19. 2024Hierarchical Classification Auxiliary Network for Time Series Forecastingcited by 26
  20. 2022Preformer: Predictive Transformer with Multi-Scale Segment-wise Correlations for Long-Term Time Series Forecastingcited by 23
  21. 2021ZEN 2.0: Continue Training and Adaption for N-gram Enhanced Text Encoderscited by 22
  22. 2022Multi-scale Attention Flow for Probabilistic Time Series Forecastingcited by 19
  23. 2023LightCTS: A Lightweight Framework for Correlated Time Series Forecastingcited by 18
  24. 2024TimeRAG: BOOSTING LLM Time Series Forecasting via Retrieval-Augmented Generationcited by 18
  25. 2024Rethinking the Power of Timestamps for Robust Time Series Forecasting: A Global-Local Fusion Perspectivecited by 13
  26. 2023Hybrid Variational Autoencoder for Time Series Forecastingcited by 8
  27. 2024Integration of Mamba and Transformer -- MAT for Long-Short Range Time Series Forecasting with Application to Weather Dynamicscited by 7
  28. 2024How Much Can Time-related Features Enhance Time Series Forecasting?cited by 6
  29. 2021Times Series Forecasting for Urban Building Energy Consumption Based on Graph Convolutional Networkcited by 5
  30. 2025Unveiling the Potential of Text in High-Dimensional Time Series Forecastingcited by 4
  31. 2023FDNet: Focal Decomposed Network for Efficient, Robust and Practical Time Series Forecastingcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 4; 2026: 4.

8 papers.

  1. 2025Diffusion Models for Time Series Forecasting: A Surveycited by 11
  2. 2025Text Reinforcement for Multimodal Time Series Forecastingcited by 6
  3. 2025UniDiff: A Unified Diffusion Framework for Multimodal Time Series Forecastingcited by 1
  4. 2025Chain-of-thought Reviewing and Correction for Time Series Question Answeringcited by 1
  5. 2026What if Tomorrow is the World Cup Final? Counterfactual Time Series Forecasting with Textual Conditionscited by 1
  6. 2026Differencing the Diffusion Trajectory toward Uncertain Components for Time Series Forecastingcited by 0
  7. 2026SCENARIODIFF: A Scenario-level Guidance Framework for Multimodal Time Series Forecasting--Extended Versioncited by 0
  8. 2026TAC-Time: Texts as Channels For Multimodal Time Series Forecastingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026DiTS: Multimodal Diffusion Transformers Are Time Series Forecasters
  2. 2024Auto-Regressive Moving Diffusion Models for Time Series Forecasting
  3. 2023Non-autoregressive Conditional Diffusion Models for Time Series Prediction
  4. 2026Enhancing few-shot time series forecasting with LLM-guided diffusion
  5. 2023Diffusion Models for Time Series Applications: A Survey
  6. 2024The Rise of Diffusion Models in Time-Series Forecasting
  7. 2023Predict, Refine, Synthesize: Self-Guiding Diffusion Models for Probabilistic Time Series Forecasting
  8. 2026Rethinking Post-Training Recipes for Multimodal Time-Series Forecasting