Metraipolis › archive › dc_59c8d6a66cdb

AdaPilot: Towards Scene-Adaptive Policy Learning for Cross-Generator Text-to-Image Quality Optimization — arXiv:2609.29517

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.29517
License (as archived)
CC0-1.0
Archived
2026-10-08T06:44:30.295Z
SHA-256 of the archived text
003cdb2b7348dc8cbd73d40ba67df9d39f28c938034a82f5d3329f1d8b00f53b
Stable URL of this page
https://metraipolis.com/documento/dc_59c8d6a66cdb

How to cite

AdaPilot: Towards Scene-Adaptive Policy Learning for Cross-Generator Text-to-Image Quality Optimization — arXiv:2609.29517 — arXiv (en), https://arxiv.org/abs/2609.29517 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_59c8d6a66cdb (Metraipolis archive dc_59c8d6a66cdb, sha256 003cdb2b7348dc8cbd73d40ba67df9d39f28c938034a82f5d3329f1d8b00f53b)
BibTeX
@misc{dc59c8d6a66cdb,
  title = {AdaPilot: Towards Scene-Adaptive Policy Learning for Cross-Generator Text-to-Image Quality Optimization — arXiv:2609.29517},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.29517},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_59c8d6a66cdb; sha256 003cdb2b7348dc8cbd73d40ba67df9d39f28c938034a82f5d3329f1d8b00f53b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.29517. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Existing methods for improving text-to-image generation quality have progressed from generator fine-tuning and prompt optimization to reinforcement learning with multi-turn visual feedback. However, existing strategies are deeply coupled with specific generators and tasks, and the learned capabilities are difficult to generalize into a universal quality optimization policy. Therefore, we propose AdaPilot, which learns a scene-adaptive, cross-generator transferable quality optimization policy by formulating multi-turn image generation as a Markov Decision Process (MDP) and optimizing it via end-to-end reinforcement learning. Specifically, AdaPilot decouples the policy from generator internals to enable cross-generator transfer, introduces scene-aware rewards that adaptively align quality assessment dimensions with task semantics, and employs process-level rewards to model the evolution trajectory of image quality. Experimental results show AdaPilot outperforms baselines in generation quality and generalization. Separate cross-generator evaluations further show that a single policy transfers zero-shot to unseen generators while maintaining positive average gains across all evaluated generators. Our project is available at this https URL .

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-25 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Computational Engineering, Finance, and Science (cs.CE).
Pagina del lavoro: https://arxiv.org/abs/2609.29517 — 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 2026-10-08T06:09:29.857Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).