Qwen-Image-PE：通过两阶段训练提升文本到图像提示增强能力
Qwen-Image-PE: Learning Visual Design for Text-to-Image Prompt Enhancement
Qwen-Image-PE 通过两阶段训练（基于标题的逆向蒸馏和分组设计奖励）提升文本到图像提示增强能力，在 Qwen-Image-Bench 上使用 Qwen-Image-2.1 时，Qwen-Image-PE-397B 相比原始请求提升 7.81 分，Qwen-Image-PE-9B 超越所有评估的 PE 基线。
Published on Oct 24
Authors:
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
Abstract
Qwen-Image-PE improves text-to-image prompts via two-stage training with caption-based reverse distillation and groupwise design rewards, scaling across data, model size, and information density.
Learning visual design is essential for prompt enhancement (PE) to turn underspecified user requests into coherent visual specifications for text-to-image (T2I) generation. Existing methods preserve user intent and improve overall visual appeal, with less explicit emphasis on fine-grained design dimensions. To address this gap, we introduce Qwen-Image-PE, which learns visual design through two-stage training. Supervised fine-tuning with Caption-Based Reverse Distillation (CRD) uses dense captions of well-designed images as design supervision. Reinforcement learning with Groupwise Design Reward (GDR) jointly compares all candidate images for the same request to refine lighting, composition, layout, and detail. On Qwen-Image-Bench with Qwen-Image-2.1, Qwen-Image-PE-397B improves over original requests by 7.81 points, while Qwen-Image-PE-9B outperforms all evaluated PE baselines. Further evaluations show gains in visual design, aesthetics, image quality, and human preference. We find that PE scales along three axes: training data, model size, and information density, with sufficient information density as a prerequisite for effective enhancement. Moreover, we demonstrate that Qwen-Image-PE generalizes to four additional T2I backbones, even surpassing native prompt enhancers on HunyuanImage-2.1 and FLUX 3 Image Klein. We release Qwen-Image-PE-9B at https://huggingface.co/Qwen/Qwen-Image-2.1-PE-T2I.
View arXiv page View PDF Add to collection
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2610.33472 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2610.33472 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2610.33472 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
来源: Hugging Face Daily Papers · huggingface.co