mlsanityData to 14 Oct 2026

SCENARIODIFF: A Scenario-level Guidance Framework for Multimodal Time Series Forecasting--Extended Version

arXiv 2608.17164 ↗First posted 17 Aug 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Textual context such as news, reports, and logs can provide valuable signals for time series forecasting, especially when future dynamics are driven by external events that are not yet visible in historical values. Existing multimodal forecasting methods often either ask large language models (LLMs) to predict numerical values directly or fuse text and time series implicitly, making contextual influence difficult to interpret and control. We propose SCENARIODIFF, a hierarchical contextual reasoning framework for multimodal time series forecasting under noisy and weakly aligned documents. SCENARIODIFF organizes contextual information into three levels: a Historical Context Agent extracts stepwise evidence from raw documents, a Scenario Agent produces a qualitative scenario description for the forecast horizon, and an Anchor Guidance Agent generates sparse anchor points for event-relevant future regions. These structured signals condition a Multimodal Diffusion Transformer, while Anchor Blended Sampling locally refines generated trajectories without retraining. Experiments on the Time-MMD benchmark show that SCENARIODIFF is especially effective in event-driven domains, demonstrating the value of explicit hierarchical scenario guidance for multimodal time series forecasting. Our full implementation is available at https://anonymous.4open.science/r/ScenarioDiff_ICDM-2C4C

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 38 of its 44 references that are in the map. First posted on arXiv 17 Aug 2026.

Builds on

38 of 44 references.

References · 38

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2020Denoising Diffusion Implicit Modelscited by 9,781
  3. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  4. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  5. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,169
  6. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 2,039
  7. 2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,676
  8. 2022Are Transformers Effective for Time Series Forecasting?cited by 1,424
  9. 2022FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecastingcited by 1,210
  10. 2022TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysiscited by 1,058
  11. 2021Blended Diffusion for Text-driven Editing of Natural Imagescited by 927
  12. 2023Time-LLM: Time Series Forecasting by Reprogramming Large Language Modelscited by 700
26 more
  1. 2023One Fits All:Power General Time Series Analysis by Pretrained LMcited by 632
  2. 2021CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputationcited by 582
  3. 2021Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecastingcited by 337
  4. 2023Long-term Forecasting with TiDE: Time-series Dense Encodercited by 332
  5. 2023TSMixer: An All-MLP Architecture for Time Series Forecastingcited by 240
  6. 2022Diffusion-based Time Series Imputation and Forecasting with Structured State Space Modelscited by 198
  7. 2025Sundial: A Family of Highly Capable Time Series Foundation Modelscited by 180
  8. 2024Are Language Models Actually Useful for Time Series Forecasting?cited by 168
  9. 2022FiLM: Frequency improved Legendre Memory Model for Long-term Time Series Forecastingcited by 147
  10. 2024CALF: Aligning LLMs for Time Series Forecasting via Cross-modal Fine-Tuningcited by 108
  11. 2024ChatTime: A Unified Multimodal Time Series Foundation Model Bridging Numerical and Textual Datacited by 100
  12. 2023Non-autoregressive Conditional Diffusion Models for Time Series Predictioncited by 78
  13. 2024Time-MMD: Multi-Domain Multimodal Dataset for Time Series Analysiscited by 74
  14. 2025Language in the Flow of Time: Time-Series-Paired Texts Weaved into a Unified Temporal Narrativecited by 40
  15. 2024Retrieval-Augmented Diffusion Models for Time Series Forecastingcited by 39
  16. 2025TimeCAP: Learning to Contextualize, Augment, and Predict Time Series Events with Large Language Model Agentscited by 37
  17. 2025Aurora: Towards Universal Generative Multimodal Time Series Forecastingcited by 29
  18. 2025Non-stationary Diffusion For Probabilistic Time Series Forecastingcited by 27
  19. 2024S2IP-LLM: Semantic Space Informed Prompt Learning with LLM for Time Series Forecastingcited by 17
  20. 2025Multi-period Learning for Financial Time Series Forecastingcited by 11
  21. 2024TEAM: Topological Evolution-aware Framework for Traffic Forecasting--Extended Versioncited by 10
  22. 2025Multimodal Conditioned Diffusive Time Series Forecastingcited by 9
  23. 2024Understanding Why Large Language Models Can Be Ineffective in Time Series Analysis: The Impact of Modality Alignmentcited by 7
  24. 2025Forecasting Clinical Risk from Textual Time Series: Structuring Narratives for Temporal AI in Healthcarecited by 5
  25. 2025Context-Aware Probabilistic Modeling with LLM for Multimodal Time Series Forecastingcited by 4
  26. 2025UniDiff: A Unified Diffusion Framework for Multimodal Time Series Forecastingcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Unlocking the Value of Text: Event-Driven Reasoning and Multi-Level Alignment for Time Series Forecasting
  2. 2026Rethinking Post-Training Recipes for Multimodal Time-Series Forecasting
  3. 2026TiMi: Empower Time Series Transformers with Multimodal Mixture of Experts
  4. 2025LLM-Integrated Bayesian State Space Models for Multimodal Time-Series Forecasting
  5. 2025BALM-TSF: Balanced Multimodal Alignment for LLM-Based Time Series Forecasting
  6. 2026WorldTS: World Modeling for Multimodal Covariate-aware Time Series Forecasting
  7. 2026OpenTSLM TeeMoE: A Unified Time-Series Language Model for Forecasting, Contextual Prediction, and Reasoning
  8. 2026CastFSR: A Fast--Slow--Reflect Agentic Reasoning Framework for Context-Aware Time Series Forecasting