Metraipolis › archive › dc_41d2d09386b4

Image Synthesis as an Intermediate for Controllable Time Series Generation — arXiv:2610.05211

Source
arXiv
Original
https://arxiv.org/abs/2610.05211
License (as archived)
CC0-1.0
Archived
2026-10-06T05:25:03.575Z
SHA-256 of the archived text
27d18dc5109696e75c253b7f0b7bde0b83c2e9e7b28fc29cebf29f87d6bb0c1c
Stable URL of this page
https://metraipolis.com/documento/dc_41d2d09386b4

How to cite

Image Synthesis as an Intermediate for Controllable Time Series Generation — arXiv:2610.05211 — arXiv, https://arxiv.org/abs/2610.05211 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_41d2d09386b4 (Metraipolis archive dc_41d2d09386b4, sha256 27d18dc5109696e75c253b7f0b7bde0b83c2e9e7b28fc29cebf29f87d6bb0c1c)
BibTeX
@misc{dc41d2d09386b4,
  title = {Image Synthesis as an Intermediate for Controllable Time Series Generation — arXiv:2610.05211},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.05211},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_41d2d09386b4; sha256 27d18dc5109696e75c253b7f0b7bde0b83c2e9e7b28fc29cebf29f87d6bb0c1c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.05211. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Semantic-driven time-series generation offers a promising way to improve downstream learning in few-shot forecasting, but directly generating numerical sequences from language often fails to preserve the intended temporal structure. We propose VisualBridge, which uses time-series plots as a visual intermediate to bridge high-level temporal semantics and numerical sequences. An MLLM first converts plotted series into structured semantic representations, enabling explicit control over temporal properties such as trend, seasonality, and volatility. We then learn a semantic editing policy with downstream forecasting rewards, allowing the generation process to favor temporal patterns that are beneficial for the target task. The resulting sequences are further modeled by a temporal VAE to produce consistent multivariate augmentations. Experiments on standard public forecasting benchmarks demonstrate that VisualBridge improves few-shot forecasting over conventional augmentation methods, with ablations validating the roles of visual semantic grounding, learned semantic control, and VAE-based generation.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.05211 — 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 2026-10-06T05:24:05.563Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).