mlsanityData to 14 Oct 2026

Chronos-2: From Univariate to Universal Forecasting

arXiv 2510.15821 ↗First posted 17 Oct 2025cs.LG cs.AI stat.MLcited by 194

Not on the map. Why?

Abstract (arXiv)

Pretrained time series models have enabled inference-only forecasting systems that produce accurate predictions without task-specific training. However, existing approaches largely focus on univariate forecasting, limiting their applicability in real-world scenarios where multivariate data and covariates play a crucial role. We present Chronos-2, a pretrained model capable of handling univariate, multivariate, and covariate-informed forecasting tasks in a zero-shot manner. Chronos-2 employs a group attention mechanism that facilitates in-context learning (ICL) through efficient information sharing across multiple time series within a group, which may represent sets of related series, variates of a multivariate series, or targets and covariates in a forecasting task. These general capabilities are achieved through training on synthetic datasets that impose diverse multivariate structures on univariate series. Chronos-2 delivers state-of-the-art performance across three comprehensive benchmarks: fev-bench, GIFT-Eval, and Chronos Benchmark II. On fev-bench, which emphasizes multivariate and covariate-informed forecasting, Chronos-2's universal ICL capabilities lead to substantial improvements over existing models. On tasks involving covariates, it consistently outperforms baselines by a wide margin. Case studies in the energy and retail domains further highlight its practical advantages. The in-context learning capabilities of Chronos-2 establish it as a general-purpose forecasting model that can be used "as is" in real-world forecasting pipelines.

Key facts. Cited by 194 papers in the map (data through 14 Oct 2026). Builds on 38 of its 65 references that are in the map. First posted on arXiv 17 Oct 2025.

Builds on

38 of 65 references.

References · 38

  1. 2017Attention Is All You Needcited by 60,361
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  4. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,703
  5. 2023Efficient Streaming Language Models with Attention Sinkscited by 2,186
  6. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 2,039
  7. 2021ViViT: A Video Vision Transformercited by 1,620
  8. 2017DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networkscited by 1,013
  9. 2019Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecastingcited by 761
  10. 2024Chronos: Learning the Language of Time Seriescited by 759
  11. 2019N-BEATS: Neural basis expansion analysis for interpretable time series forecastingcited by 753
  12. 2023Time-LLM: Time Series Forecasting by Reprogramming Large Language Modelscited by 700
26 more
  1. 2023A decoder-only foundation model for time-series forecastingcited by 692
  2. 2024Unified Training of Universal Time Series Forecasting Transformerscited by 614
  3. 2023Large Language Models Are Zero-Shot Time Series Forecasterscited by 500
  4. 2021Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecastingcited by 337
  5. 2020WeatherBench: A benchmark dataset for data-driven weather forecastingcited by 275
  6. 2021Monash Time Series Forecasting Archivecited by 267
  7. 2024GIFT-Eval: A Benchmark For General Time Series Forecasting Model Evaluationcited by 212
  8. 2025Sundial: A Family of Highly Capable Time Series Foundation Modelscited by 180
  9. 2019High-Dimensional Multivariate Forecasting with Low-Rank Gaussian Copula Processescited by 148
  10. 2022N-HiTS: Neural Hierarchical Interpolation for Time Series Forecastingcited by 146
  11. 2025TiRex: Zero-Shot Forecasting Across Long and Short Horizons with Enhanced In-Context Learningcited by 120
  12. 2023LargeST: A Benchmark Dataset for Large-Scale Traffic Forecastingcited by 118
  13. 2020Meta-learning framework with applications to zero-shot time-series forecastingcited by 70
  14. 2021Domain Adaptation for Time Series Forecasting via Attention Sharingcited by 68
  15. 2025This Time is Different: An Observability Perspective on Time Series Foundation Modelscited by 68
  16. 2025fev-bench: A Realistic Benchmark for Time Series Forecastingcited by 64
  17. 2025From Tables to Time: Extending TabPFN-v2 to Time Series Forecastingcited by 60
  18. 2023AutoGluon-TimeSeries: AutoML for Probabilistic Time Series Forecastingcited by 36
  19. 2024In-Context Fine-Tuning for Time-Series Foundation Modelscited by 35
  20. 2023BuildingsBench: A Large-Scale Dataset of 900K Buildings and Benchmark for Short-Term Load Forecastingcited by 26
  21. 2023Pushing the Limits of Pre-training for Time Series Forecasting in the CloudOps Domaincited by 23
  22. 2023LibCity: A Unified Library Towards Efficient and Comprehensive Urban Spatial-Temporal Predictioncited by 20
  23. 2025Zero-Shot Time Series Forecasting with Covariates via In-Context Learningcited by 12
  24. 2025When Does Multimodality Lead to Better Time Series Forecasting?cited by 11
  25. 2020Zero-shot and few-shot time series forecasting with ordinal regression recurrent neural networkscited by 10
  26. 2024Towards Long-Context Time Series Foundation Modelscited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 7; 2026: 187.

194 papers.

Sort:

By area:

  1. 2025Moirai 2.0: When Less Is More for Time Series Forecastingcited by 69
  2. 2026It's TIME: Towards the Next Generation of Time Series Forecasting Benchmarkscited by 33
  3. 2026Toto 2.0: Time Series Forecasting Enters the Scaling Eracited by 30
  4. 2026Timer-S1: A Billion-Scale Time Series Foundation Model with Serial Scalingcited by 28
  5. 2026TabPFN-3: Technical Reportcited by 17
  6. 2026Revisiting the Generic Transformer: Deconstructing a Strong Baseline for Time Series Foundation Modelscited by 12
  7. 2026SleepLM: Natural-Language Intelligence for Human Sleepcited by 12
  8. 2026MantisV2: Closing the Zero-Shot Gap in Time Series Classification with Synthetic Data and Test-Time Strategiescited by 11
  9. 2026QuitoBench: A High-Quality Open Time Series Forecasting Benchmarkcited by 10
  10. 2026Falcon-X: A Time Series Foundation Model for Heterogeneous Multivariate Modelingcited by 10
  11. 2026Learning the Context of Errors: Black-Box Online Adaptation of Time Series Foundation Modelscited by 9
  12. 2025Synapse: Adaptive Arbitration of Complementary Expertise in Time Series Foundational Modelscited by 7
38 more
  1. 2026TimeOmni-VL: Unified Models for Time Series Understanding and Generationcited by 7
  2. 2026GlucoFM: A Dual-Stream Foundation Model for Continuous Glucose Monitoringcited by 7
  3. 2026TFRBench: A Reasoning Benchmark for Evaluating Forecasting Systemscited by 6
  4. 2026Overcoming the Modality Gap in Context-Aided Forecastingcited by 5
  5. 2026CTF4Nuclear: Common Task Framework for Nuclear Fission and Fusion Modelscited by 5
  6. 2026KairosAgent: Agentic Time Series Forecasting with Fused Semantic Reasoningcited by 5
  7. 2026Position: Universal Time Series Foundation Models Rest on a Category Errorcited by 4
  8. 2026Position: A Dynamical Systems Perspective is Needed to Advance Time Series Modelingcited by 4
  9. 2026TempusBench: An Evaluation Framework for Time-Series Forecastingcited by 4
  10. 2026TS-ICL: A Flexible Time-Indexed Foundation Model for Time Series via In-Context Learningcited by 4
  11. 2026Rethinking Zero-Shot Time Series Classification: From Task-specific Classifiers to In-Context Inferencecited by 3
  12. 2026Is Flow Matching Just Trajectory Replay for Sequential Data?cited by 3
  13. 2026Probabilistic NDVI Forecasting from Sparse Satellite Time Series and Weather Covariatescited by 3
  14. 2026Time Series Foundation Models as Strong Baselines in Transportation Forecasting: A Large-Scale Benchmark Analysiscited by 3
  15. 2026Reasoning-Aware Training for Time Series Forecastingcited by 3
  16. 2026Toward World Modeling of Physiological Signals with Chaos-Theoretic Balancing and Latent Dynamicscited by 3
  17. 2026Why Do Time Series Models Need Long Context Windows?cited by 3
  18. 2026CardioState-JEPA: Delay-Aware Cross-Modal Learning of a Shared Cardiac Representationcited by 3
  19. 2026Into the ORBIT for Time Series: Training Regimes for Foundation Modelscited by 3
  20. 2026EXAONE Finance 1.0: An Attention-free Time Series Foundation Model for Financial Time Seriescited by 3
  21. 2026HERMES: A Unified Open-Source Framework for Realtime Multimodal Physiological Sensing, Edge AI, and Intervention in Closed-Loop Smart Healthcare Applicationscited by 2
  22. 2026CPiRi: Channel Permutation-Invariant Relational Interaction for Multivariate Time Series Forecastingcited by 2
  23. 2026Time Series Foundation Models for Energy Load Forecasting on Consumer Hardware: A Multi-Dimensional Zero-Shot Benchmarkcited by 2
  24. 2026TS-Haystack: A Multi-Task Retrieval Benchmark for Long-Context Time-Series Reasoningcited by 2
  25. 2026Time-Aware Prior Fitted Networks for Zero-Shot Forecasting with Exogenous Variablescited by 2
  26. 2026A Foundation Model for Instruction-Conditioned In-Context Time Series Taskscited by 2
  27. 2026Investigating simple target-covariate relationships for Chronos-2 and TabPFN-TScited by 2
  28. 2026MILM: Large Language Models for Multimodal Irregular Time Series with Informative Samplingcited by 2
  29. 2026Rethinking Post-Training Recipes for Multimodal Time-Series Forecastingcited by 2
  30. 2026CITRAS-FM: Tiny Time Series Foundation Model for Covariate-Informed Zero-Shot Forecastingcited by 2
  31. 2026TimeRouter: Efficient and Adaptive Routing of Time-Series Foundation Modelscited by 2
  32. 2026Zeus: Towards Tuning-Free Foundation Model for Time Series Analysiscited by 2
  33. 2026OpenMHC: Accelerating the Science of Wearable Foundation Modelscited by 2
  34. 2026DoTime: A Synthetic Benchmark Generator for Interventional and Counterfactual Time Seriescited by 2
  35. 2026Causal Analysis for Time Series Foundation Modelscited by 2
  36. 2026Beyond Numerical Time Series: A Unified Benchmark for Multimodal Forecasting with Heterogeneous Contextcited by 2
  37. 2026t0: A Time-Series Foundation Model for Forecasting with Contextcited by 2
  38. 2025Genomic Next-Token Predictors are In-Context Learnerscited by 1

50 most cited of 194 All 194 (JSON)