用多模态图像文本提升长期时间序列预测精度
Multi-Modal View Enhanced Large Vision Models for Long-Term Time Series Forecasting
- 基于趋势-季节分解与自适应回溯残差，融合多模态视图
- 在8个数据集上6个达到最优均方误差表现
- 适合需要高精度长期预测的研究者使用
时间序列通常以数值序列形式表示，也可转换为图像和文本，形成同一信号的多模态视图（MMVs）。这些视图能揭示互补模式，并使强大的预训练大模型（如大视觉模型，LVMs）可用于长期时间序列预测（LTSF）。然而，本文发现现有SOTA的基于LVM的预测器存在对“预测周期”的归纳偏置。为此，我们提出DMMV：一种基于分解的多模态视图框架，结合趋势-季节分解与新颖的回溯-残差自适应分解，整合多模态视图进行LTSF。在多个数据集上与14个SOTA模型对比，DMMV优于单视图及现有多模态基线，在8个基准数据集中的6个取得最佳均方误差（MSE）表现。代码已公开于https://github.com/D2I-Group/dmmv。
原文摘要 · Abstract (English)
Time series, typically represented as numerical sequences, can also be transformed into images and texts, offering multi-modal views (MMVs) of the same underlying signal. These MMVs can reveal complementary patterns and enable the use of powerful pre-trained large models, such as large vision models (LVMs), for long-term time series forecasting (LTSF). However, as we identified in this work, the state-of-the-art (SOTA) LVM-based forecaster poses an inductive bias towards "forecasting periods". To harness this bias, we propose DMMV, a novel decomposition-based multi-modal view framework that leverages trend-seasonal decomposition and a novel backcast-residual based adaptive decomposition to integrate MMVs for LTSF. Comparative evaluations against 14 SOTA models across diverse datasets show that DMMV outperforms single-view and existing multi-modal baselines, achieving the best mean squared error (MSE) on 6 out of 8 benchmark datasets. The code for this paper is available at: https://github.com/D2I-Group/dmmv.
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。