用原始序列做上下文学习，让时间序列预测更快更准。
Baguan-TS: A Sequence-Native In-Context Learning Model for Time Series Forecasting with Covariates
- 3D Transformer同步关注时间、变量和上下文，直接处理原始序列。
- 在多个真实能源数据集上，点预测和概率预测均显著优于基线。
- 适合需要快速适应新数据的工业级时间序列场景。
Transformers使时间序列预测实现无梯度的上下文学习（ICL），但多数方法依赖人工构造的表格特征，而端到端序列模型缺乏推理时自适应能力。本文提出统一框架Baguan-TS，将原始序列表征学习与ICL结合，采用3D Transformer在时间、变量和上下文三个维度上联合注意力。为提升高容量模型实用性，解决两大难题：(i) 校准与训练稳定性，通过目标空间检索的无特征局部校准改进；(ii) 输出过平滑问题，采用上下文过拟合策略缓解。在含协变量的公开基准测试中，Baguan-TS持续超越主流基线，赢得率最高，点预测与概率预测指标均有显著降低。跨多种真实世界能源数据集评估显示其鲁棒性强，性能提升显著。
原文摘要 · Abstract (English)
Transformers enable in-context learning (ICL) for rapid, gradient-free adaptation in time series forecasting, yet most ICL-style approaches rely on tabularized, hand-crafted features, while end-to-end sequence models lack inference-time adaptation. We bridge this gap with a unified framework, Baguan-TS, which integrates the raw-sequence representation learning with ICL, instantiated by a 3D Transformer that attends jointly over temporal, variable, and context axes. To make this high-capacity model practical, we tackle two key hurdles: (i) calibration and training stability, improved with a feature-agnostic, target-space retrieval-based local calibration; and (ii) output oversmoothing, mitigated via context-overfitting strategy. On public benchmark with covariates, Baguan-TS consistently outperforms established baselines, achieving the highest win rate and significant reductions in both point and probabilistic forecasting metrics. Further evaluations across diverse real-world energy datasets demonstrate its robustness, yielding substantial improvements.
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。