Timestral· Timestral：时间序列预测模型家族
We introduce Timestral, a family of four time series foundation models for probabilistic forecasting, with 5M, 28M, 123M, and 209M parameters. Timestral sets a new state of the art on GIFT-Eval, TIME, and fev-bench, where Timestral-209M and Timestral-123M take the top two places on every metric. We achieve this with three ingredients: Decoupled Horizon Forecasting (DHF), which separates prediction tokens from the causally encoded context, so that longer horizons are trained without masking context and each observed value is both a forecast target and context for later forecasts; a covariate framework that natively supports known-future covariates within a causal architecture; and a two-stage pipeline that pretrains exclusively on a diverse, large-scale synthetic corpus before post-training
时间序列预测模型家族 Timestral 引领新一代状态预测技术。
- 核心方法
- 采用 Decoupled Horizon Forecasting (DHF) 技术、支持未来协变量的框架和两阶段预训练与后训练的管道。
- 适合谁读
- 研究者、工程师
- 要解决的问题
- 提高时间序列预测模型的准确性，特别是在长预测范围内。
- 关键实验
- 在 GIFT-Eval、TIME 和 fev-bench 三个基准测试上进行了评估，证明了新模型的有效性。
- 主要贡献
- Timestral-209M 和 Timestral-123M 模型在 GIFT-Eval、TIME 和 fev-bench 评估指标上达到顶尖水平。
- 意义与局限
- 显著推进了时间序列预测领域的发展，特别是在处理大规模数据和长预测范围方面。模型的局限性在于预训练所需资源较大。