Metraipolis › archive › dc_77467ade66fa

Weighted Adjoint Matching: Finetuning Flow Matching Models with Time-Dependent Control Cost — arXiv:2610.20035

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.20035
License (as archived): CC0-1.0
Archived: 2026-10-15T14:23:43.500Z
SHA-256 of the archived text: a2ee103d5cfd4518f17b8216fc4eac00a2c233255bf3d867d56f00b7472427c1
Stable URL of this page: https://metraipolis.com/documento/dc_77467ade66fa

How to cite

Weighted Adjoint Matching: Finetuning Flow Matching Models with Time-Dependent Control Cost — arXiv:2610.20035 — arXiv (en), https://arxiv.org/abs/2610.20035 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_77467ade66fa (Metraipolis archive dc_77467ade66fa, sha256 a2ee103d5cfd4518f17b8216fc4eac00a2c233255bf3d867d56f00b7472427c1)

BibTeX

@misc{dc77467ade66fa,
  title = {Weighted Adjoint Matching: Finetuning Flow Matching Models with Time-Dependent Control Cost — arXiv:2610.20035},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20035},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_77467ade66fa; sha256 a2ee103d5cfd4518f17b8216fc4eac00a2c233255bf3d867d56f00b7472427c1},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20035. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We introduce Weighted Adjoint Matching, a method to finetune a pretrained diffusion or flow matching model given a differentiable reward. We derive it from Adjoint Matching, which reformulates finetuning as a stochastic optimal control problem with a quadratic control cost. Building on the observation that these models generate coarse features before fine details, we place a time-dependent weight on the control cost. A high cost early in the generative process preserves balance and diversity in the coarse features, which carry most of the semantics, while a low cost at the end lets the model increase the reward by refining details. We prove that our method is equivalent to maximizing the reward under time-varying Kullback--Leibler regularization on the generative paths. Experiments on Gaussian mixtures and on MNIST show a better tradeoff between reward and mode balance than Adjoint Matching.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (stat.ML) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.20035 — 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 2026-10-15T13:53:35.932Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).