About TS-Arena

Abstract. TS-Arena is a live benchmarking platform that evaluates Time Series Foundation Models (TSFMs) by requiring forecast submissions before ground-truth data exists — a “forecast pre-registration protocol.” This design eliminates test-set contamination and information leakage, since the evaluation target physically does not exist at submission time. The platform continuously collects forecasts from models across 186 energy-sector time series in 14 challenge definitions, scores them with MASE, and ranks them using an ELO rating system with confidence intervals.

Abstract. While Time Series Foundation Models (TSFMs) offer transformative capabilities for forecasting, they simultaneously risk triggering a fundamental evaluation crisis. This crisis is driven by information leakage due to overlapping training and test sets across different models, as well as the illegitimate transfer of global patterns to test data. While the ability to learn shared temporal dynamics represents a primary strength of these models, their evaluation on historical archives often permits the exploitation of observed global shocks, which violates the independence required for valid benchmarking. We introduce TS-Arena, a platform that restores the operational integrity of forecasting by treating the genuinely unknown future as the definitive test environment. By implementing a pre-registration mechanism on live data streams, the platform ensures that evaluation targets remain physically non-existent during inference, thereby enforcing a strict global temporal split. This methodology establishes a moving temporal frontier that prevents historical contamination and provides an authentic assessment of model generalization. Initially applied within the energy sector, TS-Arena provides a sustainable infrastructure for comparing foundation models under real-world constraints. A prototype of the platform is available at this https URL.