用统一嵌入让多个时间序列模型协同工作，零样本预测更准更快
One-Embedding-Fits-All: Efficient Zero-Shot Time Series Forecasting by a Model Zoo
- 用单一嵌入表征各模型，实现跨模型快速匹配
- 在GIFT-Eval上超越单模型，零样本表现领先
- 可动态加入新模型，适合持续迭代的部署场景
时间序列基础模型（TSFMs）的兴起显著推动了零样本预测的发展，使模型无需微调即可预测未见过的时间序列。研究表明，没有一个模型能在所有场景下通用，不同模型对特定时间模式各有偏好。这提示我们：如何利用模型间的互补能力？为此，我们提出ZooCast，通过刻画每个模型的独特预测优势，构建一个可动态选择最优模型的模型动物园。核心创新是「一嵌入适配全」范式，建立统一表示空间，使每个模型仅用一个嵌入表示，从而实现高效相似性匹配。实验表明，ZooCast在GIFT-Eval零样本预测基准上表现优异，同时保持单个TSFM的效率。在真实场景中，随着新模型陆续发布，该框架可无缝集成，以极低开销持续提升精度。
原文摘要 · Abstract (English)
The proliferation of Time Series Foundation Models (TSFMs) has significantly advanced zero-shot forecasting, enabling predictions for unseen time series without task-specific fine-tuning. Extensive research has confirmed that no single TSFM excels universally, as different models exhibit preferences for distinct temporal patterns. This diversity suggests an opportunity: how to take advantage of the complementary abilities of TSFMs. To this end, we propose ZooCast, which characterizes each model's distinct forecasting strengths. ZooCast can intelligently assemble current TSFMs into a model zoo that dynamically selects optimal models for different forecasting tasks. Our key innovation lies in the One-Embedding-Fits-All paradigm that constructs a unified representation space where each model in the zoo is represented by a single embedding, enabling efficient similarity matching for all tasks. Experiments demonstrate ZooCast's strong performance on the GIFT-Eval zero-shot forecasting benchmark while maintaining the efficiency of a single TSFM. In real-world scenarios with sequential model releases, the framework seamlessly adds new models for progressive accuracy gains with negligible overhead.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。