Metraipolis › archive › dc_bad40c33f8a5

LyapuFlow: Controlling Generative Flows with Lyapunov Feedback for Inverse Problems — arXiv:2610.04326

Source
arXiv
Original
https://arxiv.org/abs/2610.04326
License (as archived)
CC0-1.0
Archived
2026-10-06T05:30:31.190Z
SHA-256 of the archived text
f77d0ea53423426a02ff4565867e617e361dfef36bda07c349f090c8249b61e2
Stable URL of this page
https://metraipolis.com/documento/dc_bad40c33f8a5

How to cite

LyapuFlow: Controlling Generative Flows with Lyapunov Feedback for Inverse Problems — arXiv:2610.04326 — arXiv, https://arxiv.org/abs/2610.04326 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bad40c33f8a5 (Metraipolis archive dc_bad40c33f8a5, sha256 f77d0ea53423426a02ff4565867e617e361dfef36bda07c349f090c8249b61e2)
BibTeX
@misc{dcbad40c33f8a5,
  title = {LyapuFlow: Controlling Generative Flows with Lyapunov Feedback for Inverse Problems — arXiv:2610.04326},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.04326},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bad40c33f8a5; sha256 f77d0ea53423426a02ff4565867e617e361dfef36bda07c349f090c8249b61e2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04326. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Pretrained flow models are now widely used as generative priors in science and vision, where inference-time guidance enables test-time constraints without retraining. Existing methods use projection, posterior sampling, or iterative optimization of the generative trajectory. We propose LyapuFlow, an alternative based on Lyapunov feedback control. At each sampling step, LyapuFlow predicts the terminal sample towards which the current flow is evolving, and evaluates the constraint violation on this prediction. Then, we compute the minimum-norm control that satisfies a prescribed Lyapunov decrease condition. The resulting control remains inactive when the uncontrolled dynamics already reduce the constraint violation at the prescribed rate. Otherwise, it provides a corrective update within a feedback trust region that prevents the control from dominating the pretrained dynamics. We demonstrate LyapuFlow in both data and latent spaces, outperforming alternatives spanning different mechanisms for test-time constraint enforcement in scientific machine learning and image inverse problems.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2610.04326 — 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 2026-10-06T05:27:05.674Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).