Latent Inference-Time Guidance of Time Series Foundation Models
What it is
This paper introduces a method to adaptively combine forecasts from multiple Time Series Foundation Models (TSFMs) using a time-dependent latent space. It addresses the variability in individual model performance by providing a principled ensembling approach that maintains the off-the-shelf nature of foundation models.
The problem it solves
Time Series Foundation Models are sensitive to user-selected parameters (lookback, covariates, horizon), leading to variable and inconsistent forecast quality that requires manual tuning or selection of the best context.
What you could use it for
- Improving the accuracy of time series forecasting by ensembling multiple foundation models
- Mitigating the sensitivity of TSFM performance to user-selected lookback and covariates
- Creating robust forecasting pipelines that leverage complementary strengths of different TSFMs
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.