aiqbench analysis
AgenticGen introduces a novel framework that treats advertising video generation as a reasoning problem optimized directly by online business feedback loops rather than relying solely on static synthetic metrics. By decomposing the task into strategy selection and draft generation stages supervised by both performance-based and rubric-based rewards, this approach significantly improves click-through rates and conversion values in real-world TikTok experiments. This work marks a significant step toward aligning generative video models with tangible commercial outcomes through advanced reinforcement learning techniques.
AgenticGen: Reward-Guided Agentic Video Generation for Advertising
arXiv:2609.09187v1 Announce Type: cross Abstract: 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.
Abstract: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.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
arXiv:2609.09187 [cs.CV]
(or arXiv:2609.09187v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2609.09187 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 285 words
Related coverage
Links #5: 2026/07
LessWrong · Jul 31, 2026
[AINews] not much happened today
Latent Space · Sep 10, 2026
Dyna-2: General Purpose Physical Agents with Hour-Long Autonomy — Jason Ma, Dyna Robotics
Latent Space · Oct 13, 2026
What's Continual Learning, and Why Might We Expect To See It In Advanced LLM Agents?
LessWrong · Jun 12, 2026