Latent Inference-Time Guidance for TSFMs is introduced, which adaptively combines a pool of TSFM forecasts through a time-dependent latent space with independent components that comes equipped with identifiability and reconstruction guarantees, whilst maintaining the off-the-shelf aspect of foundation models.
Abstract
Time Series Foundation Models (TSFMs) currently provide state-of-the-art results in forecasting tasks. They are available out-of-the-box and rely on in-context learning to make their predictions, which makes the quality of their performance highly sensitive to the user-selected lookback, covariates, horizon and training data distributions. In practise, the quality of the forecasts are variable but complementary, which highlights the need for a principled ensembling approach, rather than selecting the best context. This paper introduces Latent Inference-Time Guidance for TSFMs, which adaptively combines a pool of TSFM forecasts through a time-dependent latent space with independent components. The framework comes equipped with identifiability and reconstruction guarantees, whilst maintaining the off-the-shelf aspect of foundation models. We provide experiments on datasets at various frequencies and from multiple domains: these show that the approach is competitive with traditional ensembling approaches.
A novel pattern-guided mixture-of-experts that expands model capacity through sparse activation and uses shallow patch similarities to constrain deep-layer routing, guiding expert specialization across heterogeneous time series and an implicit quantile network head that predicts arbitrary quantiles to characterize pred...
It is demonstrated that the way information is propagated into the model substantially affects forecast accuracy of TSFMs, and the findings suggest that current benchmarks for covariate-informed TSFMs can distort model comparisons by imposing fixed encoding strategies.
Alexander Kreusel, M. Beichter, Alexandra Nikoltchovska et al.· 0 citations
Time-series foundation models (TSFMs) have achieved strong forecasting performance across domains. However, most adaptation methods remain static. Existing all-in-one methods learn a single set of dataset-level parameter updates and apply the same adapted model to every input. As a result, they cannot adapt the model p...
Darahaas Nallagatla, Darryl C. Jacob, Pan He· 0 citations
Time series foundation models (TSFMs) are increasingly pre-trained on synthetically generated time series trajectories, where the data generating process is known. Current pre-training recipes are based on loss objectives which compare TSFM outputs to realized future values of each trajectory. We instead propose loss o...
SETTer is introduced, a transformer-based model that addresses challenges of long-term multivariate time series forecasting by incorporating novel techniques for decoupled self-attention and hybrid masking and outperforms state-of-the-art models in 88% of the scenarios.
Abraham Ezema, C. Eze, F. Ponci et al.· arXiv.org· 0 citations
Pretrained time-series foundation models (TSFMs) are evaluated as forecasters of future values, yet for sparse series many decisions depend only on which future periods contain activity. Standard benchmarks do not assess this. On five sparse datasets, we rank positions within forecast windows that contain both events a...
Daniel Schoess, F. von Wangenheim· arXiv.org· 0 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduNov 6, 2026
A machine-learning approach called CheMeleon provides a practical way to develop new drugs or other molecules with much less detailed, expensive experimentation.
Learn how Google DeepMind and GovTech Singapore uses lightweight on-device Gemma models to intercept conversational scams in private chats with zero cloud data egress.