Metraipolis › archive › dc_320af71e52de

AgenticGen: Reward-Guided Agentic Video Generation for Advertising — arXiv:2609.09187

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.09187
License (as archived): CC0-1.0
Archived: 2026-10-16T07:24:55.282Z
SHA-256 of the archived text: 317ff02db268f628601357d11f398e65ab34e266be184709284ed967f8475016
Stable URL of this page: https://metraipolis.com/documento/dc_320af71e52de

How to cite

AgenticGen: Reward-Guided Agentic Video Generation for Advertising — arXiv:2609.09187 — arXiv (en), https://arxiv.org/abs/2609.09187 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_320af71e52de (Metraipolis archive dc_320af71e52de, sha256 317ff02db268f628601357d11f398e65ab34e266be184709284ed967f8475016)

BibTeX

@misc{dc320af71e52de,
  title = {AgenticGen: Reward-Guided Agentic Video Generation for Advertising — arXiv:2609.09187},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.09187},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_320af71e52de; sha256 317ff02db268f628601357d11f398e65ab34e266be184709284ed967f8475016},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.09187. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Advertising video generation is not only a video synthesis task, but also a product-conditioned reasoning problem whose success is measured by online business metrics. Recent video foundation models can generate realistic clips from multimodal conditions, yet they do not optimize how a product should be transformed into an effective advertisement or how future generation should be improved from online business feedback. To close this loop, we propose AgenticGen, a reward-guided agentic framework that decomposes advertising video generation into two trainable reasoning stages, strategy selection and draft generation, thereby exposing optimization targets that online business feedback can supervise. AgenticGen learns a performance-based reward from accumulated online feedback and a complementary rubric-based reward aligned with human quality standards, then uses them to supervise policy optimization. DPO first moves the agentic policies toward online preferences, and GRPO further refines both stages with process and outcome rewards. Offline experiments validate the reward models and successive policy optimization. Online A/B experiments in the TikTok advertising system show that AgenticGen after DPO and GRPO improves CTR by 2.72%, CVR by 2.63%, and Advv by 9.61% over the SFT baseline.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-10 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.09187 — 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 2026-10-16T07:00:16.273Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-10?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).