Data to 7 Oct 2026

Retrieval-Augmented Diffusion Models for Time Series Forecasting

arXiv 2410.18712 ↗First posted 24 Oct 2024cs.LGcited by 35

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

also in 1 more

Diffusion models → Flow matching, time series and science

Abstract (arXiv)

While time series diffusion models have received considerable focus from many recent works, the performance of existing models remains highly unstable. Factors limiting time series diffusion models include insufficient time series datasets and the absence of guidance. To address these limitations, we propose a Retrieval- Augmented Time series Diffusion model (RATD). The framework of RATD consists of two parts: an embedding-based retrieval process and a reference-guided diffusion model. In the first part, RATD retrieves the time series that are most relevant to historical time series from the database as references. The references are utilized to guide the denoising process in the second part. Our approach allows leveraging meaningful samples within the database to aid in sampling, thus maximizing the utilization of datasets. Meanwhile, this reference-guided mechanism also compensates for the deficiencies of existing time series diffusion models in terms of guidance. Experiments and visualizations on multiple datasets demonstrate the effectiveness of our approach, particularly in complicated prediction tasks.

Key facts. Cited by 35 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 33 of its 47 references that are in the map. First posted on arXiv 24 Oct 2024.

Builds on

33 of 47 references.

References · 33

2017Attention Is All You Needcited by 59,932
2020Denoising Diffusion Probabilistic Modelscited by 20,851
2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,141
2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,126
2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 1,981
2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,654
2022Are Transformers Effective for Time Series Forecasting?cited by 1,400
2020DiffWave: A Versatile Diffusion Model for Audio Synthesiscited by 1,339

21 more

2021Improving language models by retrieving from trillions of tokenscited by 1,278
2022FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecastingcited by 1,200
2022TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysiscited by 1,030
2023iTransformer: Inverted Transformers Are Effective for Time Series Forecastingcited by 906
2017Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networkscited by 869
2019Generalization through Memorization: Nearest Neighbor Language Modelscited by 788
2019N-BEATS: Neural basis expansion analysis for interpretable time series forecastingcited by 744
2021CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputationcited by 577
2018Efficient Attention: Attention with Linear Complexitiescited by 366
2021SCINet: Time Series Modeling and Forecasting with Sample Convolution and Interactioncited by 335
2021Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecastingcited by 332
2023ReMoDiffuse: Retrieval-Augmented Motion Diffusion Modelcited by 282
2022Diffusion-based Time Series Imputation and Forecasting with Structured State Space Modelscited by 195
2023Frequency-domain MLPs are More Effective Learners in Time Series Forecastingcited by 170
2023Generative Time Series Forecasting with Diffusion, Denoise, and Disentanglementcited by 89
2023Non-autoregressive Conditional Diffusion Models for Time Series Predictioncited by 77
2024MG-TSD: Multi-Granularity Time Series Diffusion Models with Guided Learning Processcited by 31
2022DEPTS: Deep Expansion Learning for Periodic Time Series Forecastingcited by 28
2022Retrieval Based Time Series Forecastingcited by 13
2022MQRetNN: Multi-Horizon Time Series Forecasting with Retrieval Augmentationcited by 9
2021Retrieval-Augmented Transformer-XL for Close-Domain Dialog Generationcited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 13; 2026: 22.

35 papers.

Sort:

By area:

2025TS-RAG: Retrieval-Augmented Generation based Time Series Foundation Models are Stronger Zero-Shot Forecastercited by 36
2025Deep Learning and Foundation Models for Weather Prediction: A Surveycited by 15
2025Diffusion Models for Time Series Forecasting: A Surveycited by 11
2025Multimodal Conditioned Diffusive Time Series Forecastingcited by 8
2025TRACE: Grounding Time Series in Context for Multimodal Embedding and Retrievalcited by 5
2026TS-Memory: Plug-and-Play Memory for Time Series Foundation Modelscited by 5
2025RealDrive: Retrieval-Augmented Driving with Diffusion Modelscited by 3
2025LayoutRAG: Retrieval-Augmented Model for Content-agnostic Conditional Layout Generationcited by 3
2026Channel-wise Retrieval for Multivariate Time Series Forecastingcited by 3
2025Retrieval-Augmented Foundation Models for Water Level Prediction in the Evergladescited by 2
2026Enhancing Multivariate Time Series Forecasting with Global Temporal Retrievalcited by 2
2026FinStressTS: A Parametric Synthetic Benchmark for Time-Series Forecasting in Financecited by 2

23 more

2025How to Unlock Time Series Editing? Diffusion-Driven Approach with Multi-Grained Controlcited by 1
2025Learning from History: A Retrieval-Augmented Framework for Spatiotemporal Predictioncited by 1
2025Multimodal Forecasting of Sparse Intraoperative Hypotension Events Powered by Language Modelcited by 0
2025CHIME: Conditional Hallucination and Integrated Multi-scale Enhancement for Time Series Diffusion Modelcited by 0
2025EMFusion: Uncertainty-Aware Conditional Diffusion Model for Multivariate Narrow-band Exposure Forecastingcited by 0
2026Enhancing few-shot time series forecasting with LLM-guided diffusioncited by 0
2026GaiaFlow: Semantic-Guided Diffusion Tuning for Carbon-Frugal Searchcited by 0
2026Graph2TS: Structure-Controlled Time Series Generation via Quantile-Graph VAEscited by 0
2026Modular Retrieval-Augmented Generalization for Human Action Recognitioncited by 0
2026Spectral Retrieval-Augmented Time-Series Forecastingcited by 0
2026ALER-TI: Aligned Latent Embedding Retrieval for Time Series Imputationcited by 0
2026Diachronic Sample Integration: Robust Tail-Risk Estimation with Generative Modelscited by 0
2026ReDiTT: Retrieval Augmented Conditional Diffusion Transformers for Asynchronous Time Seriescited by 0
2026The Spectrum Is Not Enough: When Context Helps Time-Series Forecastingcited by 0
2026PIER: Physics-Informed Environmental Retrieval for Time-Series Modelingcited by 0
2026StrideDiffusion: Accelerating Diffusion Models for Time-series Generationcited by 0
2026Align-RAG: Alignment Is All You Need for TSFM In-Context Learningcited by 0
2026Model-agnostic Retrieval-Augmented Extended Forecasting for time seriescited by 0
2026SCENARIODIFF: A Scenario-level Guidance Framework for Multimodal Time Series Forecasting--Extended Versioncited by 0
2026When Does Retrieval Help Time-Series Forecasting?cited by 0
2026XMatch: Enhancing Covariate-Aware Time Series Forecasting through Tree-Structured Exogenous Matchingcited by 0
2026Retrieval-Augmented Diffusion Modeling for Stochastic Discount Factor Portfolioscited by 0
2026GNA: Granular Neighbor Assembly for Retrieval-Augmented Multivariate Time-Series Forecastingcited by 0

Similar, not cited

Close in wording, no citation either way