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

Xuyue Huang, Guanbo Huang, Jiasong Feng, Run Ling, Ao Ma, Xi Zhao, Yaoyu Li, Jie Gao, Ke Cao, Wei Feng, Zhanjie Zhang, Hao Wang, Jinghe HuarXiv 2026 · arXiv

TL;DR

CutAgent is a dual-agent framework that automates the creation of high-conversion e-commerce videos by decoupling creative strategy from technical execution. By introducing a structured "creative gene" and a self-refining reflection loop that learns from past production failures, it achieves a record-breaking online click-through rate (CTR) of 2.21%.

What Problem Does This Paper Address?

Creating effective e-commerce advertising videos is a balancing act between high-level creative planning (what's the hook? what are the selling points?) and low-level production (clip editing, TTS, music matching).

Current automated pipelines typically map product attributes directly to scripts or editing decisions. This "black box" approach has two major flaws:

  1. Lack of Strategy: It ignores the global advertising strategy, making it difficult to optimize the video based on commercial feedback like CTR.
  2. Fragility: These systems are error-prone—often missing key selling points or producing mismatched audio—and they lack a mechanism to "learn" from these failures to avoid repeating them in future videos.

Methods

The authors propose a Plan–Execute–Self-Evolve paradigm realized through two collaborative agents:

1. GeneAgent (The Planner)

GeneAgent acts as the creative director. It takes multi-modal product inputs (textual attributes and images) and maps them to a Creative Gene ($\widetilde{G}_i$). This "gene" is a structured representation containing:

  • Narrative Units: An ordered sequence of goals (e.g., "Attention Hook" $\rightarrow$ "Feature Demo" $\rightarrow$ "Trust Building").
  • Production Rules: Constraints for script style, voice-over tone, and music.

To optimize for conversion, GeneAgent is first fine-tuned on high-CTR historical data and then further refined using Group Relative Policy Optimization (GRPO), guided by a surrogate reward model that predicts CTR.

2. VideoAgent (The Executor & Evolver)

VideoAgent handles the heavy lifting of production. Guided by the creative gene, it:

  • Retrieves Experiences: It queries a long-term memory for "failure-correction" pairs from similar product categories to avoid past mistakes.
  • Coordinates Production: It manages shot-script planning, clip retrieval, TTS/subtitle generation, and BGM matching.
  • Self-Refinement: After rendering, an independent reflection loop evaluates the video across six dimensions (e.g., narrative quality, selling-point coverage). If defects are found, it revises the decisions and writes the verified correction back to the long-term memory.
Figure 2. Overview of CutAgent. GeneAgent takes product attributes and product images as multi-modal input Pi and maps them to a creative gene (narrative units with voice-over, music and script rules). Conditioned on the gene, VideoAgent first retrieves category-then- gene failure–correction experiences from a long-term memory E, then executes creative-gene-guided shot planning, video-clip retrieval, TTS and subtitle generation, BGM matching, and timeline rendering. Independently, a reflection loop performs evaluation, diagnosis, revision, and verification after rendering, and writes verified experiences back to E for later retrieval.

Datasets

The researchers introduced two significant resources:

  • CutAgent-70K: A massive dataset of 70,000 samples. Each sample includes product attributes, images, a "reverse-extracted" creative gene from a real ad, a pool of cleaned video clips, and real-world CTR data.
  • CutAgent-Eval: A benchmark of 5,000 products across 43 categories, used to compare CutAgent against state-of-the-art baselines.
Figure 3. Construction of CutAgent-70K and CTR-guided GeneAgent training. (a) For each product, historical ads are cleaned into a clip pool and processed by reverse gene extraction for CutAgent-70K (see Fig. 6(C1–C4)), followed by quality control; the resulting sample contains product attributes and images Pi, gene Gi, cleaned clips Mi, and CTR yi. (b) GeneAgent (Qwen3-VL-8B) is first supervised on high-CTR samples, then optimized by GRPO: K=8 candidate genes are sampled and scored by a composite reward (CTR effectiveness 0.60, format compliance 0.25, gene diversity 0.15) using a frozen LightGBM CTR-level predictor as the training-time surrogate reward (not for evaluation).

Key Insights and Ideas

  • The "Creative Gene" as an Interface: By creating a structured intermediate representation, the authors decoupled the intent of the ad from the rendering. This allows the system to optimize for high-level marketing strategies using reinforcement learning (GRPO) without getting bogged down in pixel-level details.
  • Learning from Failure: Most agents are "one-and-done." CutAgent’s use of a long-term memory for failure-correction experiences allows it to perform non-parametric self-evolution. It effectively builds a "knowledge base" of what works and what doesn't for specific product categories.
  • Surrogate CTR Rewards: Since real-world CTR is noisy and expensive to get in real-time, training the GeneAgent using a LightGBM-based surrogate reward model proved highly effective for aligning the agent with commercial goals.

Key Results

CutAgent was tested in a massive one-week online experiment involving over 10 million ad impressions.

  • Highest CTR: It achieved an online CTR of 2.21%, significantly outperforming baselines like AutoCut (1.98%) and FireRed-OpenStoryline (1.82%).
  • Human Preference: In a forced-choice A/B study, humans preferred CutAgent-generated videos 60.4% of the time.
  • Ablation Success: The "Self-Evolve" components (Reflection + Memory) provided consistent gains, raising the CTR from 1.76% (SFT only) to the final 2.21%.
Qualitative comparison showing how CutAgent creates more diverse and strategically structured narratives compared to baselines

Limitations and Future Work

  • Material Dependency: Currently, VideoAgent relies on retrieving clips from a pre-existing pool. The authors suggest that integrating generative video models (like Sora or Gen-3) could fill gaps when specific materials are missing.
  • Text Rendering: While the system handles subtitles well, the authors note that more expressive on-screen typography and spatial reasoning for text placement remain areas for improvement.
  • Broader Formats: The current focus is on short-form e-commerce videos; extending this "gene-driven" approach to long-form content or different advertising platforms is a logical next step.

Summary updated

Abstract

E-commerce advertising video creation requires both high-level creative planning and reliable multi-modal production. However, existing methods often map product attributes directly to scripts or editing decisions without explicitly modeling advertising strategies, narrative structures, and production styles. They also have limited ability to reuse past production failures. To address these limitations, we propose $\textbf{CutAgent}$, a dual-agent framework that formulates advertising video creation as a Plan--Execute--Self-Evolve process. CutAgent comprises two collaborative agents, $\textbf{GeneAgent}$ and $\textbf{VideoAgent}$, connected by a structured creative gene that bridges product understanding and video production. GeneAgent maps multi-modal product inputs---textual attributes together with product images---to creative genes that specify narrative units and production rules. To train GeneAgent, we construct $\textbf{CutAgent-70K}$ with approximately 70K samples of product attributes/images, reverse-extracted creative genes, cleaned clip pools, and real-world CTR, and apply high-CTR supervised fine-tuning followed by CTR-guided GRPO. Guided by the generated gene, VideoAgent retrieves category-then-gene failure--correction experiences from a long-term memory as constraints for shot-script planning, then coordinates material retrieval, narration, subtitles, music, and timeline rendering. Independently, a reflection loop diagnoses quality defects, revises deficient production decisions, and writes verified corrections back to memory for later retrieval. Experiments on $\textbf{CutAgent-Eval}$ show that CutAgent outperforms representative baselines in e-commerce advertising video creation, achieving the highest online CTR of $\textbf{2.21\%}$.
BibTeX
@misc{huang2026cutagent,
  author = {Xuyue Huang and Guanbo Huang and Jiasong Feng and Run Ling and Ao Ma and Xi Zhao and Yaoyu Li and Jie Gao and Ke Cao and Wei Feng and Zhanjie Zhang and Hao Wang and Jinghe Hu},
  title = {{CutAgent: Creative-Gene-Driven Planning and Experience-Guided Self-Refinement for High-CTR E-commerce Advertising Video Creation}},
  year = {2026},
  eprint = {2611.22418},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2611.22418},
}
Figure