Metraipolis › archive › dc_9ca8a0dbaf78

Beyond Trial-and-Error: Agentic Optimization for Image-to-Video Adherence — arXiv:2608.12290

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.12290
License (as archived)
CC0-1.0
Archived
2026-10-19T05:16:37.944Z
SHA-256 of the archived text
c802a4ada4e53e3c9bd68457dfab866f227df3e43331a20f3e911df064ee55be
Stable URL of this page
https://metraipolis.com/documento/dc_9ca8a0dbaf78

How to cite

Beyond Trial-and-Error: Agentic Optimization for Image-to-Video Adherence — arXiv:2608.12290 — arXiv (en), https://arxiv.org/abs/2608.12290 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9ca8a0dbaf78 (Metraipolis archive dc_9ca8a0dbaf78, sha256 c802a4ada4e53e3c9bd68457dfab866f227df3e43331a20f3e911df064ee55be)
BibTeX
@misc{dc9ca8a0dbaf78,
  title = {Beyond Trial-and-Error: Agentic Optimization for Image-to-Video Adherence — arXiv:2608.12290},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.12290},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9ca8a0dbaf78; sha256 c802a4ada4e53e3c9bd68457dfab866f227df3e43331a20f3e911df064ee55be},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.12290. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Modern black-box Image-to-Video (I2V) models offer powerful capabilities in automated content creation, yet their lack of fine-grained control and reliability presents significant challenges in professional workflows. Their inherent stochasticity causes minor variations in textual prompts or hyperparameters to yield drastically different outputs often necessitating inefficient, brute-force trial-and-error processes. To address these limitations, we introduce the ``Agentic Self-Improvement" framework, which reframes video synthesis into a closed-loop, goal-directed optimization. Our framework systematically navigates the generation parameter space using a novel two-stage approach. In the first stage, an iterative prompt optimization loop uses a multimodal Large Language Model (mLLM) to refine the input prompt. This refinement implements two automated evaluations: Davidsonian Scene Graph (DSG) queries ensure semantic adherence, and Common Mistake Questions (CMQ) for artifact detection. At the second stage, we use Bayesian optimization to efficiently co-optimize stochastic seeds and CFG scales. This search is guided by a suite of quality metrics, including the novel Video-Text Adherence (VTA) score derived from the DSG and CMQ evaluations. Our framework significantly outperforms unguided search methods: in human preference studies, videos generated via our agentic approach were strongly preferred over baseline outputs, achieving win rates up to 69\%. This work provides a practical and extensible methodology for enhancing the predictability and control of state-of-the-art video generation models, moving the field beyond speculative curiosities toward reliable, production-ready tools.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-13 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Multimedia (cs.MM).
Pagina del lavoro: https://arxiv.org/abs/2608.12290 — 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 2026-10-19T05:08:46.786Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-13?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).