Recent Transformer-based large language models (LLMs) demonstrate in-context learning ability to perform various functions based solely on the provided context, without updating model parameters. To fully utilize the in-context capabilities in time series forecasting (TSF) problems, unlike previous Transformer-based or LLM-based time series forecasting methods, we reformulate "time series forecasting tasks" as input tokens by constructing a series of (lookback, future) pairs within the tokens. This method aligns more closely with the inherent in-context mechanisms, and is more parameter-efficient without the need of using pre-trained LLM parameters. Furthermore, it addresses issues such as overfitting in existing Transformer-based TSF models, consistently achieving better performance across full-data, few-shot, and zero-shot settings compared to previous architectures.

翻译：近期基于Transformer的大型语言模型（LLM）展现出仅通过提供的上下文即可执行多种功能的上下文学习能力，而无需更新模型参数。为在时间序列预测（TSF）问题中充分利用上下文能力，与以往基于Transformer或LLM的时间序列预测方法不同，我们通过在输入标记中构建一系列（回望期，未来值）对，将“时间序列预测任务”重新表述为输入标记。该方法更贴近固有的上下文机制，且具有更高的参数效率，无需使用预训练的LLM参数。此外，它解决了现有基于Transformer的TSF模型中的过拟合等问题，在完整数据、少样本和零样本设置下，相比以往架构均能持续取得更优的性能。