Metraipolis › archive › dc_d7e8daa29ee3
Investigating the effects of the patch size in Time Series Foundation Models — arXiv:2611.02750
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.02750
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T10:58:35.637Z
- SHA-256 of the archived text
bde2f45f26325236b71766d7339707657e001efe63c425bf51e9acba042c2065- Stable URL of this page
https://metraipolis.com/documento/dc_d7e8daa29ee3
How to cite
Investigating the effects of the patch size in Time Series Foundation Models — arXiv:2611.02750 — arXiv (en), https://arxiv.org/abs/2611.02750 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d7e8daa29ee3 (Metraipolis archive dc_d7e8daa29ee3, sha256 bde2f45f26325236b71766d7339707657e001efe63c425bf51e9acba042c2065)
BibTeX
@misc{dcd7e8daa29ee3,
title = {Investigating the effects of the patch size in Time Series Foundation Models — arXiv:2611.02750},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.02750},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d7e8daa29ee3; sha256 bde2f45f26325236b71766d7339707657e001efe63c425bf51e9acba042c2065},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02750. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Patch size is a central design choice in patch-based time-series foundation models, controlling temporal resolution, sequence length, and computational cost. We study how this choice interacts with transformer architecture through a controlled comparison of encoder-only and decoder-only forecasting models trained with the same data mixture and optimization budget. Across 97 GIFT-Eval tasks, encoder-only models consistently benefit from smaller patches, whereas decoder-only performance follows a U-shaped trend, with an intermediate patch size performing best in aggregate. Dataset-level results further show that the preferred decoder patch size increases with the forecasting horizon. For encoder-only models, increasing context length improves performance across patch sizes, although the gains diminish beyond a context length of 2048. Finally, training-compute and CPU-latency Pareto analyses reveal complementary regimes: decoder-only models provide efficient low-cost alternatives, while encoder-only models attain the highest overall accuracy. These results show that the choice of patch size and architecture should be made depending on the forecasting horizon and deployment constraints. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2611.02750 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-03T10:49:38.675Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).