Time Series Forecasting Innovations

The field of time series forecasting is witnessing significant advancements with the integration of multi-modal views, large vision models, and slow-thinking language models. Researchers are exploring the potential of these innovative approaches to improve forecasting accuracy and interpretability. Notably, the use of binary cumulative encoding and retrieval-augmented time series foundation models has shown promising results. Additionally, the development of new attention mechanisms, such as XicorAttention, and calibration strategies like Socket+Plug, is enhancing the performance of existing models. Meanwhile, studies on the effectiveness of ensembling and zero-shot forecasting are providing valuable insights into the trade-offs between accuracy and computational cost.

Noteworthy papers include:

  • Multi-Modal View Enhanced Large Vision Models for Long-Term Time Series Forecasting, which proposes a novel decomposition-based multi-modal view framework for long-term time series forecasting.
  • Can Slow-thinking LLMs Reason Over Time, which investigates the potential of slow-thinking language models for time series forecasting and finds that they exhibit non-trivial zero-shot forecasting capabilities.
  • Binary Cumulative Encoding meets Time Series Forecasting, which introduces binary cumulative encoding to represent scalar targets as monotonic binary vectors and achieves state-of-the-art results on several benchmark datasets.

Sources

Multi-Modal View Enhanced Large Vision Models for Long-Term Time Series Forecasting

From Images to Signals: Are Large Vision Models Useful for Time Series Analysis?

Can Slow-thinking LLMs Reason Over Time? Empirical Studies in Time Series Forecasting

Binary Cumulative Encoding meets Time Series Forecasting

Timing is important: Risk-aware Fund Allocation based on Time-Series Forecasting

RATFM: Retrieval-augmented Time Series Foundation Model for Anomaly Detection

Univariate to Multivariate: LLMs as Zero-Shot Predictors for Time-Series Forecasting

XicorAttention: Time Series Transformer Using Attention with Nonlinear Correlation

Zero-Shot Time Series Forecasting with Covariates via In-Context Learning

Non-collective Calibrating Strategy for Time Series Forecasting

The cost of ensembling: is it always worth combining?

Built with on top of