Metraipolis › archive › dc_446ba34fceec

ZeroDiff: Zero-Shot Time Series Reconstruction via Informed-Prior Diffusion — arXiv:2609.37078

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.37078
License (as archived)
CC0-1.0
Archived
2026-10-07T17:30:26.108Z
SHA-256 of the archived text
24f7e6246dd86b9e618a50d85b810c7128a07b970df9395e732036c90f955ec5
Stable URL of this page
https://metraipolis.com/documento/dc_446ba34fceec

How to cite

ZeroDiff: Zero-Shot Time Series Reconstruction via Informed-Prior Diffusion — arXiv:2609.37078 — arXiv (en), https://arxiv.org/abs/2609.37078 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_446ba34fceec (Metraipolis archive dc_446ba34fceec, sha256 24f7e6246dd86b9e618a50d85b810c7128a07b970df9395e732036c90f955ec5)
BibTeX
@misc{dc446ba34fceec,
  title = {ZeroDiff: Zero-Shot Time Series Reconstruction via Informed-Prior Diffusion — arXiv:2609.37078},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.37078},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_446ba34fceec; sha256 24f7e6246dd86b9e618a50d85b810c7128a07b970df9395e732036c90f955ec5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.37078. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Time series modeling increasingly demands high-quality supervision, yet target observations remain scarce - exogenous inputs are broadly available, but target measurements are often unavailable due to cost, infrastructure, or accessibility constraints. Can models trained on observed locations reconstruct target time series where measurements have never been collected? We term this zero-shot time series reconstruction. A naive approach - directly mapping exogenous inputs to targets - can yield predictions at unobserved locations, but without target signals, such models fail to capture the intrinsic dynamics of the target variable, producing overly smooth outputs that underestimate extremes. This reveals systematic errors that call for explicit modeling and calibration. We propose ZeroDiff, which constructs an informed prior from exogenous variables alone, then learns to calibrate reconstruction errors through diffusion - training on observed locations and generalizing to unobserved ones. Experiments across diverse real-world datasets demonstrate significant improvements over existing approaches. Our code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.37078 — 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 2026-10-07T16:59:48.286Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).