# CommerceVibe: Learning to Design E-Commerce Creatives as Executable Visual Code via Dual-Feedback Reinforcement Learning

arXiv: https://arxiv.org/abs/2608.27893 · HTML: https://mlsanity.com/paper/2608.27893/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 32 references that are in the map.
- First posted on arXiv 28 Aug 2026.

## Builds on (most cited first, 22 of 22)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [PosterMaker: Towards High-Quality Product Poster Generation with Accurate Text Rendering](https://mlsanity.com/paper/2504.06632/) (2025)
- [AutoPoster: A Highly Automatic and Content-aware Design System for Advertising Poster Generation](https://mlsanity.com/paper/2308.01095/) (2023)
- [Graphic Design with Large Multimodal Model](https://mlsanity.com/paper/2404.14368/) (2024)
- [PosterCraft: Rethinking High-Quality Aesthetic Poster Generation in a Unified Framework](https://mlsanity.com/paper/2506.10741/) (2025)
- [Rendering-Aware Reinforcement Learning for Vector Graphics Generation](https://mlsanity.com/paper/2505.20793/) (2025)
- [A New Creative Generation Pipeline for Click-Through Rate with Stable Diffusion Model](https://mlsanity.com/paper/2401.10934/) (2024)
- [Generate E-commerce Product Background by Integrating Category Commonality and Personalized Style](https://mlsanity.com/paper/2312.13309/) (2023)
- [Towards Reliable Advertising Image Generation Using Human Feedback](https://mlsanity.com/paper/2408.00418/) (2024)
- [AeSlides: Incentivizing Aesthetic Layout in LLM-Based Slide Generation via Verifiable Rewards](https://mlsanity.com/paper/2604.22840/) (2026)
- [PosterVerse: A Full-Workflow Framework for Commercial-Grade Poster Generation with HTML-Based Scalable Typography](https://mlsanity.com/paper/2601.03993/) (2026)
- [Reason-SVG: Enhancing Structured Reasoning for Vector Graphics Generation with Reinforcement Learning](https://mlsanity.com/paper/2505.24499/) (2025)
- [AutoPP: Towards Automated Product Poster Generation and Optimization](https://mlsanity.com/paper/2512.21921/) (2025)
- [Staging E-Commerce Products for Online Advertising using Retrieval Assisted Image Generation](https://mlsanity.com/paper/2307.15326/) (2023)
- [InnoAds-Composer: Efficient Condition Composition for E-Commerce Poster Generation](https://mlsanity.com/paper/2603.05898/) (2026)
- [Efficient Optimal Selection for Composited Advertising Creatives with Tree Structure](https://mlsanity.com/paper/2103.01453/) (2021)
- [T-Stars-Poster: A Framework for Product-Centric Advertising Image Design](https://mlsanity.com/paper/2501.14316/) (2025)
- [DesignAsCode: Bridging Structural Editability and Visual Fidelity in Graphic Design Generation](https://mlsanity.com/paper/2602.17690/) (2026)
- [Seeing is Improving: Visual Feedback for Iterative Text Layout Refinement](https://mlsanity.com/paper/2603.22187/) (2026)
- [PosterReward: Unlocking Accurate Evaluation for High-Quality Graphic Design Generation](https://mlsanity.com/paper/2603.29855/) (2026)
- [GeoSVG-RL: Geometry-Aware Reinforcement Learning for Layout-Constrained Text-to-SVG Diagram Generation](https://mlsanity.com/paper/2605.25447/) (2026)
- [RefAdGen: High-Fidelity Advertising Image Generation](https://mlsanity.com/paper/2508.11695/) (2025)

## Abstract (arXiv)

High-quality e-commerce creatives are essential for presenting products and conveying marketing messages. Recent diffusion models enable scalable creative generation and produce visually compelling images, but their flattened raster outputs often contain distorted text and inconsistent product details, requiring refinement before deployment. Moreover, without explicit structure, the resulting creatives are difficult to edit and reuse, while complex design requirements remain challenging to encode as verifiable training signals. To address these challenges, we present CommerceVibe, which represents creatives as executable visual code and formulates generation as conditional HTML/CSS program synthesis. Given product images, design requirements, and product information, it produces renderable, editable, and reusable creatives. We further introduce dual-feedback reinforcement learning, in which rule-based feedback evaluates rendered programs for text readability, product visibility, and layout validity, while visual feedback from a vision-language model (VLM) assesses rendered creatives against input specifications across six perceptual and commercial dimensions. Together, these complementary feedback signals improve both constraint satisfaction and perception-dependent quality. We perform supervised fine-tuning (SFT) of Qwen3.5-9B on over 28,000 e-commerce examples, followed by dual-feedback reinforcement learning. On a 1,300-case benchmark, the optimized CommerceVibe model achieves a weighted score of 94.0/100, compared with 87.3 for the SFT-only variant, and outperforms strong external models. Blind evaluations by five e-commerce design experts further validate these improvements. CommerceVibe supports controllable, editable, and scalable e-commerce creative production.