Inherent temporal heterogeneity, such as varying sampling densities and periodic structures, has posed substantial challenges in zero-shot generalization for Time Series Foundation Models (TSFMs). Existing TSFMs predominantly rely on massive parameterization to absorb such heterogeneity, as their static tokenization and positional encoding schemes entangle diverse temporal patterns into a fixed representation space, encouraging memorization rather than adaptation. To address this limitation, we propose Kairos, a flexible and parameter-efficient TSFM that decouples temporal heterogeneity from model capacity through a novel tokenization perspective. Kairos introduces a dynamic patching tokenizer and a mixture-of-size encoding that adapt observational granularity to local information density, enabling fine-grained temporal abstraction without increasing model width or depth. In addition, we design a multi-granularity positional embedding based on dynamic rotary encodings, which conditions on instance-level spectral features and temporal structure induced by dynamic patching tokenization, allowing robust modeling of diverse temporal dependencies. Trained on a novel Predictability-Stratified Time-Series (PreSTS) corpus, Kairos achieves superior zero-shot performance with substantially fewer parameters on two mainstream benchmarks, GIFT-Eval and Time-Series-Library. The project page is at https://foundation-model-research.github.io/Kairos .

翻译：固有的时间异质性，如变化的采样密度与周期性结构，为零样本泛化的时间序列基础模型带来了重大挑战。现有TSFM主要依赖海量参数化来吸收此类异质性，因其静态分词与位置编码方案将多样的时间模式纠缠于固定表示空间，鼓励记忆而非适应。为克服此局限，我们提出Kairos——一种灵活且参数高效的TSFM，通过新颖的分词视角将时间异质性与模型容量解耦。Kairos引入动态分块分词器与混合尺度编码，使观测粒度适配局部信息密度，实现细粒度时间抽象而无需增加模型宽度或深度。此外，我们设计了基于动态旋转编码的多粒度位置嵌入，该嵌入以实例级谱特征及动态分块分词诱导的时间结构为条件，从而实现对多样时间依赖关系的鲁棒建模。在新型可预测性分层时间序列语料库上训练的Kairos，于GIFT-Eval与Time-Series-Library两大主流基准测试中，以显著更少的参数实现了卓越的零样本性能。项目页面位于https://foundation-model-research.github.io/Kairos。