Reproduce “CutAgent: Creative-Gene-Driven Planning and Experience-Guided Self-Refinement for High-CTR E-commerce Advertising Video Creation” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Xuyue Huang, Guanbo Huang, Jiasong Feng, Run Ling, Ao Ma, Xi Zhao, Yaoyu Li, Jie Gao and 5 more

CutAgent creates product ads by planning narrative beats before editing and refining production through reflection and stored corrections. Its creative gene specifies the story and production rules, while agents select footage, narration, subtitles, and music. In a one-week online test across 43 categories and 430 products, CutAgent reached 2.21% CTR versus AutoCut’s 1.98%; the bundled comparison does not isolate which components account for the difference.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.