QKCV Attention: Enhancing Time Series Forecasting with Static Categorical Embeddings for Both Lightweight and Pre-trained Foundation Models

Documenti analoghi

• TimeEmb: A Lightweight Static-Dynamic Disentanglement Framework for Time Series Forecasting
  di: Xia, Mingyuan, et al.
  Pubblicazione: (2025)
• A Mamba Foundation Model for Time Series Forecasting
  di: Ma, Haoyu, et al.
  Pubblicazione: (2024)
• TimelyGPT: Extrapolatable Transformer Pre-training for Long-term Time-Series Forecasting in Healthcare
  di: Song, Ziyang, et al.
  Pubblicazione: (2023)
• Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-Shot Forecasting of Multivariate Time Series
  di: Ekambaram, Vijay, et al.
  Pubblicazione: (2024)
• Measuring Pre-training Data Quality without Labels for Time Series Foundation Models
  di: Wen, Songkang, et al.
  Pubblicazione: (2024)