Feynman-Kac-Flow: Inference Steering of Conditional Flow Matching to an Energy-Tilted Posterior
Digital Discovery · 2026
DOI: 10.1039/d6dd00258g ↗Kurz gefasst
Fachartikel in der Zeitschrift Digital Discovery, 2026. Verfasst von Konstantin Mark und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 3.0).
Abstract (Originalsprache)
Conditional Flow Matching (CFM) represents a fast and high-quality approach to generative modelling, but methods for steering CFM during inference toward energy- or reward-tilted distributions remain underdeveloped. For diffusion models,...
Zitieren
APA
Mark, K., Galustian, L., Kovar, M. P., & Heid, E. (2026). Feynman-Kac-Flow: Inference Steering of Conditional Flow Matching to an Energy-Tilted Posterior. Digital Discovery. https://doi.org/10.1039/d6dd00258g
Vancouver
Mark K, Galustian L, Kovar MP, Heid E. Feynman-Kac-Flow: Inference Steering of Conditional Flow Matching to an Energy-Tilted Posterior. Digital Discovery. 2026. doi:10.1039/d6dd00258g
BibTeX
@article{Mark2026,
author = {Mark, Konstantin and Galustian, Leonard and Kovar, Maximilian Peter-Paul and Heid, Esther},
title = {Feynman-Kac-Flow: Inference Steering of Conditional Flow Matching to an Energy-Tilted Posterior},
journal = {Digital Discovery},
year = {2026},
doi = {10.1039/d6dd00258g}
}
EUVIMED