发表机构
Artificial Intelligence Graduate School; Ulsan National Institute of Science & Technology (UNIST); Korea University（人工智能研究生院; 蔚山科学技术院; 高丽大学）
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LTSF模型同一时间戳预测输出对齐度低的问题，提出AliO方法，引入TAM指标，可提升输出对齐度最高58.2%，同时维持或提升预测性能，增强模型可靠性。
AI 中文摘要
长期时间序列预测（LTSF）任务利用当前数据序列作为输入来预测未来序列，在天气预报、用电量规划等实际应用中愈发关键。然而，现有最先进的LTSF模型往往无法在滞后输入序列中对同一时间戳实现预测输出对齐，反而呈现出较低的输出对齐度，导致同一时间戳的预测输出出现波动，削弱了模型的可靠性。为解决该问题，我们提出了一种名为AliO（Align Outputs，对齐输出）的新方法，旨在通过减少时间和频率域中同一时间戳预测输出之间的差异，提升LTSF模型的输出对齐度。为衡量输出对齐度，我们引入了新指标TAM（Time Alignment Metric，时间对齐指标），用于量化预测输出间的对齐度，而现有指标如MSE仅能捕捉预测输出与真实值之间的距离。实验结果表明，AliO可有效提升输出对齐度，TAM指标最高提升58.2%，同时维持或提升预测性能，最高提升27.5%。这种提升的输出对齐度增强了LTSF模型的可靠性，使其更适用于实际场景。
英文摘要
Long-term Time Series Forecasting (LTSF) tasks, which leverage the current data sequence as input to predict the future sequence, have become increasingly crucial in real-world applications such as weather forecasting and planning of electricity consumption. However, state-of-the-art LTSF models often fail to achieve prediction output alignment for the same timestamps across lagged input sequences. Instead, these models exhibit low output alignment, resulting in fluctuation in prediction outputs for the same timestamps, undermining the model's reliability. To address this, we propose AliO (Align Outputs), a novel approach designed to improve the output alignment of LTSF models by reducing the discrepancies between prediction outputs for the same timestamps in both the time and frequency domains. To measure output alignment, we introduce a new metric, TAM (Time Alignment Metric), which quantifies the alignment between prediction outputs, whereas existing metrics such as MSE only capture the distance between prediction outputs and ground truths. Experimental results show that AliO effectively improves the output alignment, i.e., up to 58.2% in TAM, while maintaining or enhancing the forecasting performance (up to 27.5%). This improved output alignment increases the reliability of the LTSF models, making them more applicable in real-world scenarios.
CommentsNeurIPS 2025. 46 pages
Journal refAdvances in Neural Information Processing Systems 38 (2025), 119563-119608
