Metraipolis › archive › dc_0343e9983f10
CoRe: Coherence and Relational Alignment for Multivariate Time Series Forecasting — arXiv:2609.19670
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.19670
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-10T04:08:41.427Z
- SHA-256 of the archived text
4d1a1a82e30c77fa240305401fca3ed1dbcf75b8ded822dc26af8b31da04845c- Stable URL of this page
https://metraipolis.com/documento/dc_0343e9983f10
How to cite
CoRe: Coherence and Relational Alignment for Multivariate Time Series Forecasting — arXiv:2609.19670 — arXiv (en), https://arxiv.org/abs/2609.19670 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0343e9983f10 (Metraipolis archive dc_0343e9983f10, sha256 4d1a1a82e30c77fa240305401fca3ed1dbcf75b8ded822dc26af8b31da04845c)
BibTeX
@misc{dc0343e9983f10,
title = {CoRe: Coherence and Relational Alignment for Multivariate Time Series Forecasting — arXiv:2609.19670},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.19670},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0343e9983f10; sha256 4d1a1a82e30c77fa240305401fca3ed1dbcf75b8ded822dc26af8b31da04845c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.19670. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Direct forecasting has become a standard paradigm for multivariate time-series forecasting because it predicts the full future horizon in a single pass. However, its training objective is often still decomposed into pointwise errors such as MSE. Such objectives provide stable supervision, but they do not explicitly preserve the structure of the future trajectory: temporal coherence within each variable and relational consistency across variables can both be weakened. We propose CoRe, a model-agnostic learning objective for direct multivariate forecasting. CoRe replaces pointwise supervision with two output-space constraints: a frequency coherence loss that aligns predicted and target spectra, and a low-rank relational graph loss that matches sampled pairwise differences in a target-derived PCA subspace. The resulting objective introduces no trainable parameters and can be applied to existing forecasting backbones by changing only the loss. Experiments on standard benchmarks show that CoRe improves strong baselines, compares favorably with recent forecasting objectives, and remains effective across different backbones, datasets, and hyperparameter settings overall consistently. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.19670 — 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 2026-10-10T03:48:18.273Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).