TriTS: Time Series Forecasting from a Multimodal Perspective

TL;DR

TriTS introduces a multimodal framework for long-term time series forecasting, leveraging time, frequency, and visual modalities to improve accuracy and efficiency over existing methods.

Contribution

The paper proposes TriTS, a novel cross-modal disentanglement framework that combines time, frequency, and visual representations for enhanced long-term forecasting.

Findings

TriTS achieves state-of-the-art performance on multiple benchmarks.

It significantly reduces parameters and inference latency compared to vision-based forecasters.

The approach effectively captures complex temporal dynamics through multimodal fusion.

Abstract

Time series forecasting plays a pivotal role in critical sectors such as finance, energy, transportation, and meteorology. However, Long-term Time Series Forecasting (LTSF) remains a significant challenge because real-world signals contain highly entangled temporal dynamics that are difficult to fully capture from a purely 1D perspective. To break this representation bottleneck, we propose TriTS, a novel cross-modal disentanglement framework that projects 1D time series into orthogonal time, frequency, and 2D-vision spaces.To seamlessly bridge the 1D-to-2D modality gap without the prohibitive $O(N^2)$ computational overhead of Vision Transformers (ViTs), we introduce a Period-Aware Reshaping strategy and incorporate Visual Mamba (Vim). This approach efficiently models cross-period dependencies as global visual textures while maintaining linear computational complexity. Complementing this, we design a Multi-Resolution Wavelet Mixing (MR-WM) module for the frequency modality, which explicitly decouples non-stationary signals into trend and noise components to achieve fine-grained time-frequency localization. Finally, a streaming linear branch is retained in the time domain to anchor numerical stability. By dynamically fusing these three complementary representations, TriTS effectively adapts to diverse data contexts. Extensive experiments across multiple benchmark datasets demonstrate that TriTS achieves state-of-the-art (SOTA) performance, fundamentally outperforming existing vision-based forecasters by drastically reducing both parameter count and inference latency.