← Search

Zhongyi Pei

2 accepted papers

2026

Adapt Data to Model: Adaptive Transformation Optimization for Domain-shared Time Series Foundation Models

ICLR 2026poster

Large time series models (LTMs) have recently demonstrated powerful capabilities for universal forecasting. However, these models still struggle to address the variety and nonstationarity of time series, resulting in an unsatisfying balance between forecasting performance and generalizability. Inste…

Cited by 0SourcecodeScholar
2026

TiMi: Empower Time Series Transformers with Multimodal Mixture of Experts

ICML 2026poster

Multimodal time series forecasting has garnered significant attention for its potential to provide more robust and accurate predictions than traditional single-modality models by leveraging rich information inherent in other modalities. However, due to fundamental challenges in modality alignment, e…

Cited by 0SourceScholar