mlsanityData to 7 Oct 2026

Retrieval Augmented Time Series Forecasting

arXiv 2411.08249 ↗First posted 12 Nov 2024cs.LG cs.AIcited by 40

On the map: Retrieval-augmented generation → Time series and agricultural RAG

Abstract (arXiv)

Retrieval-augmented generation (RAG) is a central component of modern LLM systems, particularly in scenarios where up-to-date information is crucial for accurately responding to user queries or when queries exceed the scope of the training data. The advent of time-series foundation models (TSFM), such as Chronos, and the need for effective zero-shot forecasting performance across various time-series domains motivates the question: Do benefits of RAG similarly carry over to time series forecasting? In this paper, we advocate that the dynamic and event-driven nature of time-series data makes RAG a crucial component of TSFMs and introduce a principled RAG framework for time-series forecasting, called Retrieval Augmented Forecasting (RAF). Within RAF, we develop efficient strategies for retrieving related time-series examples and incorporating them into forecast. Through experiments and mechanistic studies, we demonstrate that RAF indeed improves the forecasting accuracy across diverse time series domains and the improvement is more significant for larger TSFM sizes.

Key facts. Cited by 40 papers in the map (data through 7 Oct 2026). Placed in Time series and agricultural RAG, one of 15 directions in Retrieval-augmented generation. Builds on 47 of its 67 references that are in the map. First posted on arXiv 12 Nov 2024.

Builds on

47 of 67 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024RATSF: Empowering Customer Service Volume Management through Retrieval-Augmented Time-Series Forecasting7 Mar

Other references · 46

  1. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  2. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  3. 2020Scaling Laws for Neural Language Modelscited by 6,411
  4. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
  5. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,581
  6. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,141
  7. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,126
  8. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 1,981
  9. 2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,654
  10. 2022Are Transformers Effective for Time Series Forecasting?cited by 1,400
  11. 2021Improving language models by retrieving from trillions of tokenscited by 1,278
  12. 2020Leveraging Passage Retrieval with Generative Models for Open Domain Question Answeringcited by 1,203
34 more
  1. 2022FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecastingcited by 1,200
  2. 2016Instance Normalization: The Missing Ingredient for Fast Stylizationcited by 1,187
  3. 2020ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERTcited by 1,160
  4. 2017DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networkscited by 1,004
  5. 2023iTransformer: Inverted Transformers Are Effective for Time Series Forecastingcited by 906
  6. 2019Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecastingcited by 748
  7. 2024Chronos: Learning the Language of Time Seriescited by 712
  8. 2019Latent Retrieval for Weakly Supervised Open Domain Question Answeringcited by 691
  9. 2023A decoder-only foundation model for time-series forecastingcited by 650
  10. 2019Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecastingcited by 606
  11. 2024Unified Training of Universal Time Series Forecasting Transformerscited by 588
  12. 2024MOMENT: A Family of Open Time-series Foundation Modelscited by 453
  13. 2021Transformers Can Do Bayesian Inferencecited by 338
  14. 2021Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecastingcited by 332
  15. 2023Can Generalist Foundation Models Outcompete Special-Purpose Tuning? Case Study in Medicinecited by 319
  16. 2021Monash Time Series Forecasting Archivecited by 256
  17. 2024Foundation Models for Time Series Analysis: A Tutorial and Surveycited by 234
  18. 2023TSMixer: An All-MLP Architecture for Time Series Forecastingcited by 232
  19. 2023Query Rewriting for Retrieval-Augmented Large Language Modelscited by 211
  20. 2022A Survey on Retrieval-Augmented Text Generationcited by 157
  21. 2023TimeGPT-1cited by 149
  22. 2021AdaRNN: Adaptive Learning and Forecasting of Time Seriescited by 120
  23. 2023Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecastingcited by 102
  24. 2023ForecastPFN: Synthetically-Trained Zero-Shot Forecastingcited by 102
  25. 2019Deep Factors for Forecastingcited by 86
  26. 2020Meta-learning framework with applications to zero-shot time-series forecastingcited by 70
  27. 2021Domain Adaptation for Time Series Forecasting via Attention Sharingcited by 68
  28. 2021Baleen: Robust Multi-Hop Reasoning at Scale via Condensed Retrievalcited by 63
  29. 2024Retrieval & Fine-Tuning for In-Context Tabular Modelscited by 58
  30. 2020Augmenting Transformers with KNN-Based Composite Memory for Dialoguecited by 53
  31. 2024TOTEM: TOkenized Time Series EMbeddings for General Time Series Analysiscited by 37
  32. 2024Agentic Retrieval-Augmented Generation for Time Series Analysiscited by 13
  33. 2020Zero-shot and few-shot time series forecasting with ordinal regression recurrent neural networkscited by 10
  34. 2024Retrieval-Augmented Mining of Temporal Logic Specifications from Datacited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 9; 2026: 31.

40 papers.

Sort:

By area:

  1. 2025TS-RAG: Retrieval-Augmented Generation based Time Series Foundation Models are Stronger Zero-Shot Forecastercited by 36
  2. 2025Improving Time Series Forecasting via Instance-aware Post-hoc Revisioncited by 14
  3. 2026MemCast: Memory-Driven Time Series Forecasting with Experience-Conditioned Reasoningcited by 8
  4. 2026Not All Retrievals are Useful: Cross-Attention for Input-Aware RAG in Time Series Forecastingcited by 4
  5. 2026Unlocking the Value of Text: Event-Driven Reasoning and Multi-Level Alignment for Time Series Forecastingcited by 4
  6. 2026Channel-wise Retrieval for Multivariate Time Series Forecastingcited by 3
  7. 2026Post-Training in Time Series Foundation Models: A Unifying Frameworkcited by 3
  8. 2025Large Language models for Time Series Analysis: Techniques, Applications, and Challengescited by 2
  9. 2025Retrieval-Augmented Foundation Models for Water Level Prediction in the Evergladescited by 2
  10. 2026Enhancing Multivariate Time Series Forecasting with Global Temporal Retrievalcited by 2
  11. 2026MEMTS: Internalizing Domain Knowledge via Parameterized Memory for Retrieval-Free Domain Adaptation of Time Series Foundation Modelscited by 2
  12. 2026Retrieval Mechanisms Surpass Long-Context Scaling in Time Series Forecastingcited by 2
28 more
  1. 2026Stationarity-Aware Retrieval-Augmented Time Series Forecastingcited by 2
  2. 2026Semantics-Enhanced Retrieval-Augmented Time Series Forecastingcited by 2
  3. 2026Factorize to Generalize: Retrieval-Guided Invariant-Dynamic Decomposition for Time Series Forecastingcited by 1
  4. 2025QuiZSF: A Retrieval-Augmented Framework for Zero-Shot Time Series Forecastingcited by 0
  5. 2025STRATA-TS: Selective Knowledge Transfer for Urban Time Series Forecasting with Retrieval-Guided Reasoningcited by 0
  6. 2025RAxSS: Retrieval-Augmented Sparse Sampling for Explainable Variable-Length Medical Time Series Classificationcited by 0
  7. 2025Resolution-Aware Retrieval Augmented Zero-Shot Forecastingcited by 0
  8. 2025AutoFed: Personalized Federated Traffic Prediction via Adaptive Promptcited by 0
  9. 2026Dual-Prototype Disentanglement: A Context-Aware Enhancement Framework for Time Series Forecastingcited by 0
  10. 2026RAAP: Retrieval-Augmented Affordance Prediction with Cross-Image Action Alignmentcited by 0
  11. 2026Beyond Extrapolation: Knowledge Utilization Paradigm with Bidirectional Inspiration for Time Series Forecastingcited by 0
  12. 2026Dynamic TMoE: A Drift-Aware Dynamic Mixture of Experts Framework for Non-Stationary Time Series Forecastingcited by 0
  13. 2026Spectral Retrieval-Augmented Time-Series Forecastingcited by 0
  14. 2026ALER-TI: Aligned Latent Embedding Retrieval for Time Series Imputationcited by 0
  15. 2026ReDiTT: Retrieval Augmented Conditional Diffusion Transformers for Asynchronous Time Seriescited by 0
  16. 2026The Spectrum Is Not Enough: When Context Helps Time-Series Forecastingcited by 0
  17. 2026PIER: Physics-Informed Environmental Retrieval for Time-Series Modelingcited by 0
  18. 2026Align-RAG: Alignment Is All You Need for TSFM In-Context Learningcited by 0
  19. 2026TSDS-Toolbox: A Toolbox for Measuring Time-Series Dataset Similaritycited by 0
  20. 2026Model-agnostic Retrieval-Augmented Extended Forecasting for time seriescited by 0
  21. 2026Which Histories Matter for Time Series Forecasting? Learning Predictive Relevance with Future Supervisioncited by 0
  22. 2026HypLTSF: A Hyperbolic Geometric View of Multi-Scale Hierarchies for Long-Term Time Series Forecastingcited by 0
  23. 2026When Does Retrieval Help Time-Series Forecasting?cited by 0
  24. 2026Knowledge-Graph-Augmented Chronos-2 for HEC-RAS Surrogate Forecastingcited by 0
  25. 2026WorldTS: World Modeling for Multimodal Covariate-aware Time Series Forecastingcited by 0
  26. 2026XMatch: Enhancing Covariate-Aware Time Series Forecasting through Tree-Structured Exogenous Matchingcited by 0
  27. 2026GNA: Granular Neighbor Assembly for Retrieval-Augmented Multivariate Time-Series Forecastingcited by 0
  28. 2026SteerCast: Retrieval-Based Latent Steering for Decoder-Only Time Series Forecastingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026TS-RAG: Retrieval Augmented Generation for Time Series Forecasting
  2. 2024TimeRAG: BOOSTING LLM Time Series Forecasting via Retrieval-Augmented Generation
  3. 2024TimeRAF: Retrieval-Augmented Foundation model for Zero-shot Time Series Forecasting
  4. 2025Retrieval Augmented Time Series Forecasting
  5. 2025Retrieval-augmented Large Language Models for Financial Time Series Forecasting
  6. 2026Retrieval-Augmented Generation with Covariate Time Series
  7. 2026Retrieval Is Not Enough: Refreshing Memory for Frozen Time-Series Forecasters
  8. 2026TFGformer: Multivariate Time Series Forecasting via Time-Frequency Graph Learning and Covariate Fusion