Position: Time-Series Foundation Models Require Explicit Domain-Level Benchmarks

Md Asif Bin Syed, Md Younus Ahamed, Azmine Toushik Wasi
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:172221-172247, 2026.

Abstract

Time series foundation models (TSFMs) have demonstrated strong performance on established benchmarks such as GIFT-Eval, Monash, and TSFM-Bench. However, these benchmarks pool datasets from many domains with uneven representation, which can obscure performance within specific application areas such as healthcare, finance, nature, retail, and transport. The necessity for domain-specific evaluation arises from the inherent structural diversity of time series data: clinical records often feature irregular sampling and informative missingness; financial sequences are characterized by high noise and stochastic trajectories; and environmental data, such as energy and weather, are governed by deterministic physical laws and strong seasonal hierarchies. Motivated by this heterogeneity, we argue that TSFMs require explicit domain-specific benchmarks so practitioners can reliably assess a model’s utility within their own application area. This is because cross-domain differences in data generation, sampling irregularity, and nonstationarity under concept drift fundamentally shape forecasting difficulty and failure modes. As a result, strong performance on aggregated leaderboards may not translate to reliable deployment within a specific domain. To test this, we evaluated seven TSFMs across 72 datasets from six domains (healthcare, finance, energy, nature, transport, and retail) and found substantial cross-domain variability. These findings confirm that global benchmark scores can be misleading and that domain-aware evaluations are essential for trustworthy TSFM selection.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-syed26a, title = {Position: Time-Series Foundation Models Require Explicit Domain-Level Benchmarks}, author = {Syed, Md Asif Bin and Ahamed, Md Younus and Wasi, Azmine Toushik}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {172221--172247}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/syed26a/syed26a.pdf}, url = {https://proceedings.mlr.press/v306/syed26a.html}, abstract = {Time series foundation models (TSFMs) have demonstrated strong performance on established benchmarks such as GIFT-Eval, Monash, and TSFM-Bench. However, these benchmarks pool datasets from many domains with uneven representation, which can obscure performance within specific application areas such as healthcare, finance, nature, retail, and transport. The necessity for domain-specific evaluation arises from the inherent structural diversity of time series data: clinical records often feature irregular sampling and informative missingness; financial sequences are characterized by high noise and stochastic trajectories; and environmental data, such as energy and weather, are governed by deterministic physical laws and strong seasonal hierarchies. Motivated by this heterogeneity, we argue that TSFMs require explicit domain-specific benchmarks so practitioners can reliably assess a model’s utility within their own application area. This is because cross-domain differences in data generation, sampling irregularity, and nonstationarity under concept drift fundamentally shape forecasting difficulty and failure modes. As a result, strong performance on aggregated leaderboards may not translate to reliable deployment within a specific domain. To test this, we evaluated seven TSFMs across 72 datasets from six domains (healthcare, finance, energy, nature, transport, and retail) and found substantial cross-domain variability. These findings confirm that global benchmark scores can be misleading and that domain-aware evaluations are essential for trustworthy TSFM selection.} }
Endnote
%0 Conference Paper %T Position: Time-Series Foundation Models Require Explicit Domain-Level Benchmarks %A Md Asif Bin Syed %A Md Younus Ahamed %A Azmine Toushik Wasi %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-syed26a %I PMLR %P 172221--172247 %U https://proceedings.mlr.press/v306/syed26a.html %V 306 %X Time series foundation models (TSFMs) have demonstrated strong performance on established benchmarks such as GIFT-Eval, Monash, and TSFM-Bench. However, these benchmarks pool datasets from many domains with uneven representation, which can obscure performance within specific application areas such as healthcare, finance, nature, retail, and transport. The necessity for domain-specific evaluation arises from the inherent structural diversity of time series data: clinical records often feature irregular sampling and informative missingness; financial sequences are characterized by high noise and stochastic trajectories; and environmental data, such as energy and weather, are governed by deterministic physical laws and strong seasonal hierarchies. Motivated by this heterogeneity, we argue that TSFMs require explicit domain-specific benchmarks so practitioners can reliably assess a model’s utility within their own application area. This is because cross-domain differences in data generation, sampling irregularity, and nonstationarity under concept drift fundamentally shape forecasting difficulty and failure modes. As a result, strong performance on aggregated leaderboards may not translate to reliable deployment within a specific domain. To test this, we evaluated seven TSFMs across 72 datasets from six domains (healthcare, finance, energy, nature, transport, and retail) and found substantial cross-domain variability. These findings confirm that global benchmark scores can be misleading and that domain-aware evaluations are essential for trustworthy TSFM selection.
APA
Syed, M.A.B., Ahamed, M.Y. & Wasi, A.T.. (2026). Position: Time-Series Foundation Models Require Explicit Domain-Level Benchmarks. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:172221-172247 Available from https://proceedings.mlr.press/v306/syed26a.html.

Related Material