Metraipolis › archive › dc_897fc7cd457b
MeanFlowAdvantage: Stable Reward Fine-Tuning for Few-Step Average-Velocity Generators — arXiv:2609.37670
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.37670
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T16:24:01.669Z
- SHA-256 of the archived text
70ea5c47cd497230655b24f7e96559e3317cce45961405fdcaf1b99cc69f14da- Stable URL of this page
https://metraipolis.com/documento/dc_897fc7cd457b
How to cite
MeanFlowAdvantage: Stable Reward Fine-Tuning for Few-Step Average-Velocity Generators — arXiv:2609.37670 — arXiv (en), https://arxiv.org/abs/2609.37670 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_897fc7cd457b (Metraipolis archive dc_897fc7cd457b, sha256 70ea5c47cd497230655b24f7e96559e3317cce45961405fdcaf1b99cc69f14da)
BibTeX
@misc{dc897fc7cd457b,
title = {MeanFlowAdvantage: Stable Reward Fine-Tuning for Few-Step Average-Velocity Generators — arXiv:2609.37670},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.37670},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_897fc7cd457b; sha256 70ea5c47cd497230655b24f7e96559e3317cce45961405fdcaf1b99cc69f14da},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.37670. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
MeanFlow enables efficient few-step generation by predicting interval-average velocities, but this representation creates a mismatch for reward fine-tuning: existing advantage-based objectives are typically defined on instantaneous velocities or equivalent $x_0$-space predictions, whereas inference directly uses the learned average-velocity map. We introduce MeanFlowAdvantage, a signed advantage-weighted least-squares objective for average-velocity generators. Our key construction uses a shared, detached MeanFlow derivative correction to express the reward objective in prediction space while making rollout and reference regularization exact penalties on the average-velocity network deployed at inference. The resulting formulation preserves MeanFlow's native few-step sampler and provides a direct mechanism for transferring reward improvements to the deployed flow map. On SD3.5-Medium, MeanFlowAdvantage improves all eight reported metrics over the matched four-step MeanFlowNFT baseline and, with only four NFEs, matches or exceeds the 40-step DiffusionNFT baseline on six of eight metrics. The same objective also transfers to DNA promoter design, where it supports both teacher-free on-policy RL for a generator defined on a manifold and teacher-guided reward-graded distillation, with the latter yielding the lowest one-step Sei profile MSE among the compared configurations. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI) ; Computational Engineering, Finance, and Science (cs.CE); Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.37670 — 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 2026-10-07T15:54:53.643Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).