mlsanityData to 7 Oct 2026

ForecastPFN: Synthetically-Trained Zero-Shot Forecasting

arXiv 2311.01933 ↗First posted 3 Nov 2023cs.LGcited by 102

Not on the map. Why?

Abstract (arXiv)

The vast majority of time-series forecasting approaches require a substantial training dataset. However, many real-life forecasting applications have very little initial observations, sometimes just 40 or fewer. Thus, the applicability of most forecasting methods is restricted in data-sparse commercial applications. While there is recent work in the setting of very limited initial data (so-called `zero-shot' forecasting), its performance is inconsistent depending on the data used for pretraining. In this work, we take a different approach and devise ForecastPFN, the first zero-shot forecasting model trained purely on a novel synthetic data distribution. ForecastPFN is a prior-data fitted network, trained to approximate Bayesian inference, which can make predictions on a new time series dataset in a single forward pass. Through extensive experiments, we show that zero-shot predictions made by ForecastPFN are more accurate and faster compared to state-of-the-art forecasting methods, even when the other methods are allowed to train on hundreds of additional in-distribution data points.

Key facts. Cited by 102 papers in the map (data through 7 Oct 2026). Builds on 37 of its 63 references that are in the map. First posted on arXiv 3 Nov 2023.

Builds on

37 of 63 references.

References · 37

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2022FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awarenesscited by 3,611
  4. 2020Longformer: The Long-Document Transformercited by 3,362
  5. 2014Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modelingcited by 2,341
  6. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,141
  7. 2020Reformer: The Efficient Transformercited by 1,658
  8. 2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,654
  9. 2022FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecastingcited by 1,200
  10. 2017DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networkscited by 1,004
  11. 2017Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networkscited by 869
  12. 2020Efficient Transformers: A Surveycited by 784
25 more
  1. 2022FourCastNet: A Global Data-driven High-resolution Weather Model using Adaptive Fourier Neural Operatorscited by 748
  2. 2019N-BEATS: Neural basis expansion analysis for interpretable time series forecastingcited by 744
  3. 2022TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Secondcited by 643
  4. 2018Deep learning for time series classification: a reviewcited by 575
  5. 2023Large Language Models Are Zero-Shot Time Series Forecasterscited by 491
  6. 2020Time Series Forecasting With Deep Learning: A Surveycited by 457
  7. 2020AutoGluon-Tabular: Robust and Accurate AutoML for Structured Datacited by 368
  8. 2021Transformers Can Do Bayesian Inferencecited by 338
  9. 2017Multitask learning and benchmarking with clinical time series datacited by 287
  10. 2019Financial Time Series Forecasting with Deep Learning : A Systematic Literature Review: 2005-2019cited by 270
  11. 2023When Do Neural Nets Outperform Boosted Trees on Tabular Data?cited by 218
  12. 2017A Multi-Horizon Quantile Recurrent Forecastercited by 211
  13. 2023Large Models for Time Series and Spatio-Temporal Data: A Survey and Outlookcited by 109
  14. 2023Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecastingcited by 102
  15. 2020Meta-learning framework with applications to zero-shot time-series forecastingcited by 70
  16. 2023PFNs4BO: In-Context Learning for Bayesian Optimizationcited by 70
  17. 2018Transfer learning for time series classificationcited by 60
  18. 2023Statistical Foundations of Prior-Data Fitted Networkscited by 59
  19. 2023Efficient Bayesian Learning Curve Extrapolation using Prior-Data Fitted Networkscited by 37
  20. 2023AutoGluon-TimeSeries: AutoML for Probabilistic Time Series Forecastingcited by 36
  21. 2020Global Models for Time Series Forecasting: A Simulation Studycited by 21
  22. 2021ClimART: A Benchmark Dataset for Emulating Atmospheric Radiative Transfer in Weather and Climate Modelscited by 21
  23. 2022SDWPF: A Dataset for Spatial Dynamic Wind Power Forecasting Challenge at KDD Cup 2022cited by 21
  24. 2019Machine Learning vs Statistical Methods for Time Series Forecasting: Size Matterscited by 10
  25. 2020Zero-shot and few-shot time series forecasting with ordinal regression recurrent neural networkscited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 39; 2025: 31; 2026: 32.

102 papers.

Sort:

By area:

  1. 2024Chronos: Learning the Language of Time Seriescited by 712
  2. 2024Unified Training of Universal Time Series Forecasting Transformerscited by 588
  3. 2024Timer: Generative Pre-trained Transformers Are Large Time Series Modelscited by 198
  4. 2024Deep Time Series Models: A Comprehensive Survey and Benchmarkcited by 191
  5. 2024ForecastBench: A Dynamic Benchmark of AI Forecasting Capabilitiescited by 104
  6. 2024Approaching Human-Level Forecasting with Language Modelscited by 102
  7. 2024TuneTables: Context Optimization for Scalable Prior-Data Fitted Networkscited by 47
  8. 2025Do-PFN: In-Context Learning for Causal Effect Estimationcited by 46
  9. 2024Is Mamba Capable of In-Context Learning?cited by 45
  10. 2024Retrieval Augmented Time Series Forecastingcited by 40
  11. 2024Towards a General Time Series Anomaly Detector with Adaptive Bottlenecks and Dual Adversarial Decoderscited by 37
  12. 2024LLMs learn governing principles of dynamical systems, revealing an in-context neural scaling lawcited by 32
38 more
  1. 2024Towards a General Time Series Forecasting Model with Unified Representation and Adaptive Transfercited by 31
  2. 2024In-Context Fine-Tuning for Time-Series Foundation Modelscited by 31
  3. 2024Large language models can be zero-shot anomaly detectors for time series?cited by 30
  4. 2025TimePFN: Effective Multivariate Time Series Forecasting with Synthetic Datacited by 27
  5. 2024In-Context Freeze-Thaw Bayesian Optimization for Hyperparameter Optimizationcited by 26
  6. 2024Zero-shot forecasting of chaotic systemscited by 25
  7. 2024Drift-Resilient TabPFN: In-Context Learning Temporal Distribution Shifts on Tabular Datacited by 25
  8. 2026Timer-S1: A Billion-Scale Time Series Foundation Model with Serial Scalingcited by 24
  9. 2024Towards Urban General Intelligence: A Review and Outlook of Urban Foundation Modelscited by 23
  10. 2024Empowering Time Series Analysis with Foundation Models: A Comprehensive Surveycited by 23
  11. 2024In-context Time Series Predictorcited by 21
  12. 2025How Can Time Series Analysis Benefit From Multiple Modalities? A Survey and Outlookcited by 20
  13. 2024In-Context Data Distillation with TabPFNcited by 19
  14. 2025Position: The Future of Bayesian Prediction Is Prior-Fittedcited by 19
  15. 2024Mamba4Cast: Efficient Zero-Shot Time Series Forecasting with State Space Modelscited by 17
  16. 2025BLAST: Balanced Sampling Time Series Corpus for Universal Forecasting Modelscited by 16
  17. 2025TempoPFN: Synthetic Pre-training of Linear RNNs for Zero-shot Time Series Forecastingcited by 16
  18. 2024On Catastrophic Inheritance of Large Foundation Modelscited by 15
  19. 2024Generative Pretrained Hierarchical Transformer for Time Series Forecastingcited by 15
  20. 2024TabEBM: A Tabular Data Augmentation Method with Distinct Class-Specific Energy-Based Modelscited by 15
  21. 2025Out-of-Distribution Generalization in Time Series: A Surveycited by 15
  22. 2025Context parroting: A simple but tough-to-beat baseline for foundation models in scientific machine learningcited by 14
  23. 2024Only the Curve Shape Matters: Training Foundation Models for Zero-Shot Multivariate Time Series Forecasting through Next Curve Shape Predictioncited by 11
  24. 2024ViTime: Foundation Model for Time Series Forecasting Powered by Vision Intelligencecited by 11
  25. 2025Diffusion Models for Time Series Forecasting: A Surveycited by 11
  26. 2026Relational In-Context Learning via Synthetic Pre-training with Structural Priorcited by 11
  27. 2025FeDaL: Federated Dataset Learning for General Time Series Foundation Modelscited by 10
  28. 2025Rethinking Evaluation in the Era of Time Series Foundation Models: (Un)known Information Leakage Challengescited by 10
  29. 2026Learning Posterior Predictive Distributions for Node Classification from Synthetic Graph Priorscited by 10
  30. 2025Towards Foundation Models for Zero-Shot Time Series Anomaly Detection: Leveraging Synthetic Data and Relative Context Discrepancycited by 9
  31. 2026From Zero to Hero: Advancing Zero-Shot Foundation Models for Tabular Outlier Detectioncited by 8
  32. 2025In-context Learning of Evolving Data Streams with Tabular Foundational Modelscited by 7
  33. 2025In-Context Learning of Stochastic Differential Equations with Foundation Inference Modelscited by 7
  34. 2024LaT-PFN: A Joint Embedding Predictive Architecture for In-context Time-series Forecastingcited by 6
  35. 2026Interventional Time Series Priors for Causal Foundation Modelscited by 6
  36. 2024BTS: Building Timeseries Dataset: Empowering Large-Scale Building Analyticscited by 5
  37. 2024Towards Foundation Time Series Model: To Synthesize Or Not To Synthesize?cited by 4
  38. 2024Beyond Data Scarcity: A Frequency-Driven Framework for Zero-Shot Forecastingcited by 4

50 most cited of 102 All 102 (JSON)