Evaluating Accuracy and Probabilistic Reliability of Zero-Shot Time Series Foundation Models

The study reveals that while TSFMs outperform statistical methods and supervised models, they are subject to a fundamental trade-off between point accuracy and probabilistic reliability. It comes from a paper posted to arXiv on 23 September 2026. Time Series Foundation Models (TSFMs) promise a paradigm shift toward zero-shot forecasting by eliminating task-specific training. However, existing works often overlook trade-offs between predictive accuracy and probabilistic calibration. This paper presents a benchmark study of six TSFMs evaluated on energy, traffic, and financial datasets. We contrast their performance against statistical baselines and a supervised DL model. Specifically, xLSTM architectures provide robust probabilistic calibration across horizons. In contrast, patch-based transformers offer competitive accuracy but face calibration issues at long horizons, while transformer-based models exhibit context saturation points for optimal zero-shot reasoning. These findings offer evidence-based guidance for balancing generalization and uncertainty quantification in real-world deployments.

Why it counts

The study reveals that while TSFMs outperform statistical methods and supervised models, they are subject to a fundamental trade-off between point accuracy and probabilistic reliability.

Sources

Primary source: primary source

What is not known

This brief does not claim independent replication. Claims that appear only on X and not in the primary source stay unknown.