Chronos-2: From Univariate to Universal Forecasting

2025-10-17

AI Summary

Existing pretrained time-series models predominantly focus on univariate forecasting and struggle to generalize to realistic multivariate settings with exogenous covariates. To address this limitation, we propose the first universal time-series foundation model supporting zero-shot forecasting for univariate, multivariate, and covariate-augmented scenarios. Our approach synthesizes diverse multivariate structures via data augmentation and introduces a group-wise attention mechanism to enable efficient cross-series information sharing. Crucially, we pioneer the extension of in-context learning to multivariate and covariate-dependent forecasting tasks. The model follows a pretrain-and-zero-shot-inference paradigm and achieves state-of-the-art performance across three major benchmarks—FEV-Bench, GIFT-Eval, and Chronos Benchmark II—particularly excelling in covariate-aware forecasting, where it significantly outperforms prior methods. These results demonstrate its strong generalization capability and practical applicability to real-world time-series forecasting.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: (Large) Language Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

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.

Problem

• Extends pretrained models from univariate to universal forecasting tasks
• Enables zero-shot multivariate and covariate-informed time series prediction
• Uses group attention for in-context learning across related series

Innovation

• Pretrained model handles univariate multivariate covariate forecasting
• Group attention enables efficient cross-series information sharing
• Training on synthetic datasets achieves universal forecasting capabilities