Chronos-2: From Univariate to Universal Forecasting

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ansari, Abdul Fatir, Shchur, Oleksandr, Küken, Jaris, Auer, Andreas, Han, Boran, Mercado, Pedro, Rangapuram, Syama Sundar, Shen, Huibin, Stella, Lorenzo, Zhang, Xiyuan, Goswami, Mononito, Kapoor, Shubham, Maddix, Danielle C., Guerron, Pablo, Hu, Tony, Yin, Junming, Erickson, Nick, Desai, Prateek Mutalik, Wang, Hao, Rangwala, Huzefa, Karypis, George, Wang, Yuyang, Bohlke-Schneider, Michael
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915559803715584
author Ansari, Abdul Fatir
Shchur, Oleksandr
Küken, Jaris
Auer, Andreas
Han, Boran
Mercado, Pedro
Rangapuram, Syama Sundar
Shen, Huibin
Stella, Lorenzo
Zhang, Xiyuan
Goswami, Mononito
Kapoor, Shubham
Maddix, Danielle C.
Guerron, Pablo
Hu, Tony
Yin, Junming
Erickson, Nick
Desai, Prateek Mutalik
Wang, Hao
Rangwala, Huzefa
Karypis, George
Wang, Yuyang
Bohlke-Schneider, Michael
author_facet Ansari, Abdul Fatir
Shchur, Oleksandr
Küken, Jaris
Auer, Andreas
Han, Boran
Mercado, Pedro
Rangapuram, Syama Sundar
Shen, Huibin
Stella, Lorenzo
Zhang, Xiyuan
Goswami, Mononito
Kapoor, Shubham
Maddix, Danielle C.
Guerron, Pablo
Hu, Tony
Yin, Junming
Erickson, Nick
Desai, Prateek Mutalik
Wang, Hao
Rangwala, Huzefa
Karypis, George
Wang, Yuyang
Bohlke-Schneider, Michael
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2510_15821
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Chronos-2: From Univariate to Universal Forecasting
Ansari, Abdul Fatir
Shchur, Oleksandr
Küken, Jaris
Auer, Andreas
Han, Boran
Mercado, Pedro
Rangapuram, Syama Sundar
Shen, Huibin
Stella, Lorenzo
Zhang, Xiyuan
Goswami, Mononito
Kapoor, Shubham
Maddix, Danielle C.
Guerron, Pablo
Hu, Tony
Yin, Junming
Erickson, Nick
Desai, Prateek Mutalik
Wang, Hao
Rangwala, Huzefa
Karypis, George
Wang, Yuyang
Bohlke-Schneider, Michael
Machine Learning
Artificial Intelligence
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.
title Chronos-2: From Univariate to Universal Forecasting
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2510.15821