Metraipolis › archive › dc_95ebd0ea02f3

CommerceGen-Studio: An End-to-End Framework for Commercial-Grade Main-Image Generation — arXiv:2610.20603

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.20603
License (as archived): CC0-1.0
Archived: 2026-10-15T21:57:31.625Z
SHA-256 of the archived text: 227e88fabd42a2ec5a62f146957378696fc72662241f8d93a89d0647ea5b108f
Stable URL of this page: https://metraipolis.com/documento/dc_95ebd0ea02f3

How to cite

CommerceGen-Studio: An End-to-End Framework for Commercial-Grade Main-Image Generation — arXiv:2610.20603 — arXiv (en), https://arxiv.org/abs/2610.20603 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_95ebd0ea02f3 (Metraipolis archive dc_95ebd0ea02f3, sha256 227e88fabd42a2ec5a62f146957378696fc72662241f8d93a89d0647ea5b108f)

BibTeX

@misc{dc95ebd0ea02f3,
  title = {CommerceGen-Studio: An End-to-End Framework for Commercial-Grade Main-Image Generation — arXiv:2610.20603},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20603},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_95ebd0ea02f3; sha256 227e88fabd42a2ec5a62f146957378696fc72662241f8d93a89d0647ea5b108f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20603. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.20603 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T21:38:29.553Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).