发表机构
KAIST（韩国科学技术院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有在线时间序列预测框架难以长期适应与泛化的问题，提出基于LLM的CoSPOT框架，采用组合频谱提示实现高效在线适应，在多类在线预测场景中表现优异。
AI 中文摘要
为解决时间序列的序列性与演化特性，在线时间序列预测（OTSF）任务已在多个领域得到广泛研究。现有研究聚焦于通过基于内存缓冲区的检索策略适应非平稳环境，但我们发现这类框架在长期适应方面存在困难，且无法泛化到未见过的模式。为此，我们提出CoSPOT，这是一种基于大语言模型（LLM）的在线时间序列预测框架，以预训练LLM作为核心在线预测器，其动机源于LLM强大的少样本能力。为实现高效的在线适应，CoSPOT保持LLM冻结状态，并采用基于频域基的组合频谱提示，通过输入的整体分布引导模型，从而大幅减少在线阶段更新的参数数量。具体而言，CoSPOT将时间序列分解为频域基，并根据其幅值组合相应的频谱基提示，使未见过的模式可表示为已学习基提示的新组合。我们在真实世界数据集上开展的大量实验表明，CoSPOT在具有挑战性的在线场景中（包括延长的在线阶段和存在显著分布偏移的跨数据集设置）具有优越性和实用性。我们的代码可在该https链接获取。
英文摘要
To address the sequential and evolving nature of time series, the Online Time Series Forecasting (OTSF) task has been extensively studied in multiple domains. Existing research focuses on adapting to non-stationary environments by employing memory buffer-based retrieval strategies. However, we observe that such frameworks struggle with long-term adaptation and fail to generalize to unseen patterns. To this end, we introduce CoSPOT, an LLM-based online time series forecasting framework that leverages a pre-trained LLM as the backbone online forecaster, motivated by its strong few-shot capabilities. For efficient online adaptation, CoSPOT keeps the LLM frozen and employs compositional spectral prompts grounded in frequency-domain bases to guide the model with the overall distribution of the input, thereby substantially reducing the number of parameters updated during the online phase. Specifically, CoSPOT decomposes time series into frequency bases and composes the corresponding spectral basis prompts according to their amplitudes, allowing unseen patterns to be represented as new combinations of learned basis prompts. Our extensive experiments on real-world datasets demonstrate the superiority and practicality of CoSPOT across challenging online scenarios, including extended online phases and cross-dataset settings with substantial distribution shifts. Our code is available at https://github.com/seungyoon-Choi/CoSPOT.
CommentsCIKM 2026
