PulseAugur
中
实时 07:37:40
English(EN) Scale-Invariant Training for Time Series Foundation Models

新的时间序列基础模型应对零样本预测和尺度不变性

研究人员推出了一种新颖的多变量时间序列基础模型 Timer-M1,该模型通过学习时间与关系原语来设计用于零样本预测。该模型利用基于原语的数据合成和预训练流程,并结合了门控二维 Transformer 块来实现动态跨变量注意力。同时,另一项研究提出了面向时间序列基础模型的尺度不变性训练 (ScaleIn),以解决尺度污染训练 (ScaleCon) 问题,在这种问题中,尺度较大的序列会对模型产生不成比例的影响。ScaleIn 旨在通过确保梯度不被高尺度序列主导来提高收敛性和预测准确性,并在各种基准测试和架构中显著降低了误差。 AI

影响 时间序列基础模型的这些进展可能带来更稳健、更准确的跨领域预测,影响金融、气候科学和资源管理等领域。

排序理由 arXiv 上发表了两篇关于时间序列基础模型新方法的论文。

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

新的时间序列基础模型应对零样本预测和尺度不变性

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
arXiv 上发表了两篇关于时间序列基础模型新方法的论文。
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
paper, model release
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
46 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [2]

  1. arXiv cs.AI TIER_1 English(EN) · Haoran Zhang, Haixuan Liu, Xingjian Su, Yong Liu, Zhi Chen, Yuxuan Wang, Jianmin Wang, Mingsheng Long ·

    Timer-M1: 通过学习基本要素实现的多变量时间序列基础模型

    arXiv:2610.11734v1 Announce Type: cross Abstract: We introduce Timer-M1, a pretrained multivariate time series foundation model that learns with primitives for zero-shot forecasting. Across domains, time series share elementary temporal and relational patterns, termed primitives,…

  2. arXiv cs.AI TIER_1 English(EN) · Ignacy Stepka, Willa Potosnak, Kin G. Olivares, Artur Dubrawski ·

    Scale-Invariant Training for Time Series Foundation Models

    arXiv:2610.07324v1 Announce Type: cross Abstract: Time series foundation models (TSFMs) are trained on large collections of time series datasets that span various morphologies and domains. This setting exposes models to series whose scales -- typical magnitudes of their values --…