#machine learning
Feb 2026
TiMi: Empower Time Series Transformers with Multimodal Mixture of Experts
This paper introduces a Multimodal Mixture-of-Experts (MMoE) module as a lightweight plug-in to empower Transformer-based time series models for multimodal forecasting, eliminating the need for explicit representation-level alignment.
Jiafeng Lin, Yuxuan Wang, Huakun Luo et al.
· arXiv.org · 1 citation