{"slug": "commercegen-studio-an-end-to-end-framework-for-commercial-grade-main-image", "title": "CommerceGen-Studio: An End-to-End Framework for Commercial-Grade Main-Image Generation", "summary": "A team of researchers posted CommerceGen-Studio to arXiv on 14 Oct 2026, an end-to-end framework for generating commercial-grade main images for e-commerce SKUs that runs in production at millions of main images per day with a latency of a few seconds per image. The framework combines a Prompt-Eval-Refine-Loop Agent (PERL-Agent), Layout Thinker Optimization (LTO), and SKU-DiT trained under Reinforce-Distill-Reinforce (RDR) progressive training, conditioning jointly on the white-background product image, the trademark crop, and the structured layout. The authors also released CommerceGen-Bench, on which the framework scores highest overall among evaluated open-source baselines on both editing tracks and beats several larger closed-source models at Instruction Following and Product Integration.", "body_md": "# Computer Science > Computer Vision and Pattern Recognition\n\n  [Submitted on 14 Oct 2026]\n\n# Title:CommerceGen-Studio: An End-to-End Framework for Commercial-Grade Main-Image Generation\n\n[View PDF](http://arxiv.org/pdf/2610.20603v1)\n\n[HTML (experimental)](https://arxiv.org/html/2610.20603v1)\n\nAbstract:We present CommerceGen-Studio, an end-to-end framework for generating commercial-grade main images for e-commerce Stock Keeping Units (SKUs). Our framework is organized along a three-stage pipeline of information extraction, semantic rewriting, and multi-condition generation: Prompt-Eval-Refine-Loop Agent (PERL-Agent) distills heterogeneous SKU metadata into concise selling-point text lists through an iterative closed loop of VLM and LLM roles scored by a reference-free rubric. Layout Thinker Optimization (LTO) aligns the rewriter with both caption and rendered-image quality via alternating domain-specific policy optimization. SKU-DiT, trained under Reinforce-Distill-Reinforce (RDR) progressive training, conditions jointly on the white-background product image, the trademark crop, and the structured layout to hold subject fidelity, trademark integrity, and character-level text accuracy together. The framework runs in production at a scale of millions of main images per day with a latency of a few seconds per image, which we believe makes it among the first systems of this kind to operate at commercial scale. We also release CommerceGen-Bench, a benchmark of hard cases collected from live traffic across all product categories. On it, our framework achieves the highest overall scores among the evaluated open-source baselines on both editing tracks, with and without prompt enhancement. It also matches much larger closed-source models, and beats several of them at Instruction Following and Product Integration - the two properties that determine whether an image can be published.\n    \n\n### Additional Features\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer \n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers \n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps \n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations \n\n*(*[What are Smart Citations?](https://www.scite.ai/))\n# Code, Data and Media Associated with this Article\n\nalphaXiv \n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers \n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub \n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub \n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face \n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast \n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))\n# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower \n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender \n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))\n# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth 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.\n\nHave an idea for a project that will add value for arXiv's community? [**Learn more about arXivLabs**](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/commercegen-studio-an-end-to-end-framework-for-commercial-grade-main-image", "canonical_source": "http://arxiv.org/abs/2610.20603v1", "published_at": "2026-10-15 14:35:56+00:00", "updated_at": "2026-10-15 14:53:13.342667+00:00", "lang": "en", "topics": ["computer-vision", "generative-ai", "ai-research", "large-language-models", "ai-products"], "entities": ["CommerceGen-Studio", "arXiv", "PERL-Agent", "Layout Thinker Optimization", "SKU-DiT", "Reinforce-Distill-Reinforce", "CommerceGen-Bench"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/commercegen-studio-an-end-to-end-framework-for-commercial-grade-main-image", "markdown": "https://wpnews.pro/news/commercegen-studio-an-end-to-end-framework-for-commercial-grade-main-image.md", "text": "https://wpnews.pro/news/commercegen-studio-an-end-to-end-framework-for-commercial-grade-main-image.txt", "jsonld": "https://wpnews.pro/news/commercegen-studio-an-end-to-end-framework-for-commercial-grade-main-image.jsonld"}}