TimeSeek 提出多任务多域预训练，统一时间序列预测、异常检测与分类
TimeSeek: Multi-Task and Multi-Domain Pretraining for Generalized Time Series Analytics
这个 TimeSeek 把预测、异常检测、分类三个任务塞进了一次预训练里，架构上按领域自适应提取特征，不再只靠堆 Transformer 层。做时间序列的工程师可以对比一下 TimesFM、Chronos 这类单任务模型，看 few-shot 和 zero-shot 成绩是否真能省掉多套模型的维护成本。
TimeSeek 是一种新的通用时间序列分析模型，针对现有时间序列基础模型（TSFM）的两个痛点：只靠堆叠 Transformer 层扩参数、预训练局限于单一任务（如仅预测或仅异常检测）。模型核心是 Multi-Domain Feature Extractor，通过渐进式多尺度建模和自适应槽位通道建模，按领域特征自适应捕捉时间依赖与通道相关性。
在任务层面，TimeSeek 采用多任务协同训练，通过任务定制化与跨任务训练同时利用任务特异性和共性，在预训练阶段统一预测、异常检测、分类三类任务。作者称 TimeSeek 在大规模时间序列数据上预训练后，于 few-shot 和 zero-shot 两种设定下在这三类任务上均取得 SOTA 成绩，同时保持较好的推理效率。
与主流 TSFM 相比，TimeSeek 的差异点在于不依赖单纯加深网络，而是在架构和训练目标上分别处理域异质性与任务异质性，适合需要一套模型覆盖多个下游场景的团队。
TimeSeek: Multi-Task and Multi-Domain Pretraining for Generalized Time Series Analytics
Time series analytics has been widely applied across various domains. While existing time series foundation models (TSFMs) have demonstrated promising results, they still face two major challenges: 1) Most TSFMs mainly rely on stacking Transformer layers to increase parameter size, lacking the ability to adaptively model complex time series patterns from multi-domain data. 2) Current TSFMs are primarily pre-trained for specific tasks, e.g., forecasting or anomaly detection, and a few models support multiple tasks, but they fail to fully leverage task specificity and commonality, resulting in suboptimal performance. To address these challenges, we propose TimeSeek, a novel generalized time series analytics model that adaptively models time series patterns from multi-domain data and unifies multiple tasks in pre-training. Conceptually, TimeSeek is guided by selective information sharing under domain and task heterogeneity. From the perspective of multi-domain data, we propose a Multi-Domain Feature Extractor that adaptively captures temporal dependencies and channel correlations based on domain-specific characteristics, including progressive multi-scale modeling and adaptive slot channel modeling. From the perspective of the diversity of tasks, we propose multi-task collaborative training, which leverages task specificity and task commonality through task customization and cross-task training, effectively unifying multiple tasks in pre-training and better adapting to diverse downstream scenarios. After pre-training on large-scale time series data, TimeSeek achieves state-of-the-art performance in forecasting, anomaly detection, and classification tasks across both few-shot learning and zero-shot situations, together with favorable inference efficiency.
本条情报归属 AITOP「模型」核心追踪栏目。在跨区域产业观察坐标中，该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」研究主线，反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括：TimeSeek、时间序列、TimeSer 基础模型、异常检测，读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明：本文原始线索采编自「arXiv cs.LG」（@Peng Chen, Yihang Wang, Zhongwen Rao, Yang Shu, Yuying Qiu, Yuxuan Chen, Lujia Pan, Bin Yang, Chenjuan Guo），经 AITOP 多维评分管线结构化提炼。
