研究量化 MOMENT、Chronos-2 等 4 个时序基础模型的采样频率不变性
Quantifying the Sampling Frequency Invariance of Time-Series Foundation Models for Sensor Data Applications
搞传感器和时序预测的人该看看：论文拿 MOMENT、Chronos-2、Moirai-MoE、TimesFM 四个模型实测，告诉你降采样到什么程度性能才开始崩，还给了 FSI、FDI、CDI 三个现成指标。做边缘设备省电部署时能直接照着评估。
arXiv 论文对 4 个主流时序基础模型（TSFM）——MOMENT、Chronos-2、Moirai-MoE 和 TimesFM——进行了采样频率（SF）变化的压力测试，覆盖传感器数据应用中因设备省电模式等导致的采样率变化场景。
研究设计了 3 个评估维度：嵌入偏移、预测性能退化和下游任务性能退化，分别用三个新指标 FSI、FDI 和 CDI 量化。核心发现是模型行为取决于任务所需信号带宽与原生采样率之间的余量，而非采样频率数值本身；也就是说只要采样率仍保留任务关键带宽，TSFM 就能保持有效。
这是首个把采样频率不变性作为 TSFM 评估核心的研究。论文将 SF 不变性重构为（信号 × TSFM）配对的可测量属性，为开发者在传感器部署中判断何时能安全降低采样频率提供了依据，同时区分了性能下降中信息损失与模型能力局限的比例。
Quantifying the Sampling Frequency Invariance of Time-Series Foundation Models for Sensor Data Applications
Time-Series Foundation Models (TSFMs) are increasingly used in sensor applications, where sampling frequency (SF) varies across sensors, devices, and operating modes (e.g., power saving). Ideally, TSFMs must remain effective during such SF changes, provided the SF retains task-critical bandwidth. In this paper, we stress-test 4 prominent TSFMs: MOMENT, Chronos-2, Moirai-MoE, and TimesFM, to quantify their SF-invariance and investigate whether their behavior tracks the resulting loss of information bandwidth rather than the SF value itself. Our stress-test probes 3 dimensions: (i) the shift in embeddings, (ii) the degradation in forecasting performance, and (iii) the degradation in downstream performance, quantified by three novel metrics: FSI, FDI, and CDI, respectively. To our knowledge, this is the first study to place SF-invariance at the core of TSFM evaluations. Our results characterize: (i) how SF-invariance depends on the margin between signal's native-SF and the bandwidth its task requires, (ii) how much of that invariance the choice of TSFM explains, and (iii) how much of TSFM's performance degradation under a lower SF reflects information loss rather than a limitation of their abilities. These findings reframe SF-invariance as a measurable, signal-aware property of the (signal X TSFM) pair, and guide practitioners on when TSFMs display SF-invariance.
This report is categorized under AITOP's "Papers" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include MOMENT, Chronos-2, TimesFM, 时序基础模型.
Source attribution: curated from arXiv cs.AI (@Sourish Gunesh Dhekane, Abinaya Kumar, Torsten Woertwein, Yan-Bo Lin, Thomas Ploetz, Shabnam Ghaffarzadegan).