InnoAds-Composer：面向电商海报生成的高效条件组合
InnoAds-Composer: Efficient Condition Composition for E-Commerce Poster Generation
把主体、字形、风格三类条件以词元形式注入单阶段扩散模型，并按层与时间步的重要性只路由到最敏感位置，这一「条件词元稀疏路由」思路可直接迁移到任何多条件可控生成任务（如多参考图定制化、布局+文本+风格联合控制），避免条件拼接带来的二次开销。
原文依据 we perform importance analysis over layers and timesteps and route each condition only to the most responsive positions
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论电商产品海报生成旨在通过呈现主体、文字与设计风格，自动合成一张能有效传达产品信息的单一图像。近期具备细粒度且高效可控性的扩散模型推动了产品海报合成的发展，但这些方法通常依赖多阶段流水线，且对主体、文字与风格的同时控制仍研究不足。此类朴素的多阶段流水线还暴露出三个问题：主体保真度差、文字不准确、风格不一致。为解决这些问题，我们提出 InnoAds-Composer，一个单阶段框架，可在主体、字形与风格上实现高效的三条件控制词元。为缓解朴素三条件词元拼接带来的二次开销，我们对层与时间步进行重要性分析，并将每个条件仅路由到最敏感的位置，从而缩短活跃词元序列。此外，为提升中文文字渲染的准确性，我们设计了文本特征增强模块（TFEM），融合来自字形图像与字形裁剪块的特征。为支持训练与评测，我们还构建了一个高质量的电商产品海报数据集与基准，这是首个同时包含主体、文字与风格条件的数据集。大量实验表明，InnoAds-Composer 显著优于现有产品海报方法，且未明显增加推理延迟。
英文原文摘要 1510 字符
E-commerce product poster generation aims to automatically synthesize a single image that effectively conveys product information by presenting a subject, text, and a designed style. Recent diffusion models with fine-grained and efficient controllability have advanced product poster synthesis, yet they typically rely on multi-stage pipelines, and simultaneous control over subject, text, and style remains underexplored. Such naive multi-stage pipelines also show three issues: poor subject fidelity, inaccurate text, and inconsistent style. To address these issues, we propose InnoAds-Composer, a single-stage framework that enables efficient tri-conditional control tokens over subject, glyph, and style. To alleviate the quadratic overhead introduced by naive tri-conditional token concatenation, we perform importance analysis over layers and timesteps and route each condition only to the most responsive positions, thereby shortening the active token sequence. Besides, to improve the accuracy of Chinese text rendering, we design a Text Feature Enhancement Module (TFEM) that integrates features from both glyph images and glyph crops. To support training and evaluation, we also construct a high-quality e-commerce product poster dataset and benchmark, which is the first dataset that jointly contains subject, text, and style conditions. Extensive experiments demonstrate that InnoAds-Composer significantly outperforms existing product poster methods without obviously increasing inference latency.