研究人员正在探索用于时间序列分析的新架构，重点关注多元数据。一项研究发现，更简单的状态空间模型（SSM），特别是 S4D 变体，在分类任务中优于更复杂的基于 Mamba 的模型，并引入了 MS4 和 MS4N 等轻量级修改。同时，已开发出用于异构多元时间序列的 Falcon-X 基础模型，将变量解耦到潜在原型空间，以更好地对齐和建模复杂交互。此外，还发布了一个名为 FactoryNet 的新大规模数据集，以促进工业时间序列基础模型的发展，该数据集具有统一的模式，用于跨实体转移和异常检测。 AI
影响 时间序列建模和基础模型的进步可以改善复杂工业环境中的预测和异常检测。
排序理由 arXiv 上发布了多篇研究论文，详细介绍了用于时间序列分析的新模型和数据集。
AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →
arXiv:2605.27406v1 Announce Type: new Abstract: Structured state space models (SSMs) have recently emerged as a promising foundation for sequence modeling, with Mamba-based architectures demonstrating strong performance through input-dependent state transitions, albeit at conside…
arXiv:2605.27286v1 Announce Type: cross Abstract: Time series foundation models (TSFMs) are transforming the forecasting paradigm through large-scale cross-domain pretraining. However, most existing TSFMs remain univariate, and recent efforts to enable cross-variate modeling stil…
Time series foundation models (TSFMs) are transforming the forecasting paradigm through large-scale cross-domain pretraining. However, most existing TSFMs remain univariate, and recent efforts to enable cross-variate modeling still operate directly within the raw variate space. T…
arXiv:2605.09081v3 Announce Type: replace-cross Abstract: We introduce the first universal pretraining corpus for industrial time-series data: FactoryNet. 51M datapoints across 23k end-to-end task executions (13.3k real, 9.8k synthetic) on six embodiments, unified by a shared sch…