arXiv Machine Learning July 13, 2026 By Willa Potosnak, Malcolm Wolff, Mengfei Cao, Ruijun Ma, Tatiana Konstantinova, Dmitry Efimov, Michael W. Mahoney, Boris Oreshkin, Kin G. Olivares

Forking-Sequences: Statistically and Computationally Efficient Multi-Horizon Forecasting with Reduced Volatility

Read the original on arXiv Machine Learning →

arXiv:2510. 04487v5 Announce Type: replace Abstract: While accuracy is a critical requirement for time series forecasting, an equally important desideratum is reasonable forecast volatility across forecast creation dates (FCDs).

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

arXiv AI

Aug 10

Seeking SOTA: Time-Series Forecasting Must Adopt Taxonomy-Specific Evaluation to Dispel Illusory Gains

arXiv:2603. 15506v2 Announce Type: replace-cross Abstract: We argue that the current practice of evaluating AI/ML time-series forecasting models, predominantly on benchmarks characterized by strong, persistent periodicities and seasonalities, obscures real progress by overlooking the performance of efficient classical methods.

By Raeid Saqur, Christoph Bergmeir, Blanka Horvath, Daniel Schmidt, Frank Rudzicz, Terry Lyons

benchmarks

arXiv Machine Learning

Aug 4

AROpt: An Optimization Method for Autoregressive Time Series Forecasting

arXiv:2602. 02288v3 Announce Type: replace Abstract: Current time-series forecasting models are primarily based on transformer-style neural networks.

By Zheng Li, Jerry Cheng, Huanying Gu

llms benchmarks

arXiv AI

Jun 29

Deployment-Side Adaptiveness in Multi-Horizon Volatility Forecasting

arXiv:2606. 27688v1 Announce Type: cross Abstract: In financial forecasting, predictive performance depends not only on which model is trained, but also on how the trained model is deployed.

By Riku Green, Zahraa S. Abdallah, Telmo M Silva Filho

arXiv Machine Learning

6d ago

Regime-Gated Residual Mixture-of-Experts for Cross-Sectional Volatility Forecasting

arXiv:2608. 12251v1 Announce Type: cross Abstract: Financial volatility is regime dependent, yet incorporating regime information into neural networks can also destabilize training.

By Junyi Ye, Gargi Vijay Borde

arXiv Machine Learning

Jul 2

StateFlow: Dual-State Recurrent Modeling for Long-Horizon Time Series Forecasting

arXiv:2607. 00197v1 Announce Type: new Abstract: Long-horizon multivariate time series forecasting (LTSF) remains challenging due to non-stationarity, regime shifts, and error accumulation.

By Haroon Gharwi, Yue Dai, Kai Shu

llms benchmarks

arXiv Machine Learning

Jun 2

Optimizing accuracy and diversity: a multi-task approach to forecast combinations

arXiv:2310. 20545v3 Announce Type: replace Abstract: We present a multi-task optimization approach based on a deep learning architecture for time series forecasting.

By Giovanni Felici, Antonio M. Sudoso

benchmarks

Related stories

Seeking SOTA: Time-Series Forecasting Must Adopt Taxonomy-Specific Evaluation to Dispel Illusory Gains

AROpt: An Optimization Method for Autoregressive Time Series Forecasting

Deployment-Side Adaptiveness in Multi-Horizon Volatility Forecasting

Regime-Gated Residual Mixture-of-Experts for Cross-Sectional Volatility Forecasting

StateFlow: Dual-State Recurrent Modeling for Long-Horizon Time Series Forecasting

Optimizing accuracy and diversity: a multi-task approach to forecast combinations