PosterIQ：面向海报理解与生成的设计视角基准
PosterIQ: A Design Perspective Benchmark for Poster Understanding and Generation
为视觉语言模型与扩散生成器提供海报理解与生成的细粒度诊断基准，可用于评测模型在版面解析、文字-图像对应、字体感知与构图可控生成上的能力，并暴露商业模型作为自动评分器不敏感的问题。
原文依据 We present PosterIQ, a design-driven benchmark for poster understanding and generation
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论我们提出 PosterIQ，一个面向海报理解与生成的设计驱动基准，其标注涵盖构图结构、字体层级与语义意图。它包含 7,765 个图像-标注实例和 822 个生成提示，覆盖真实、专业与合成案例。为连接视觉设计认知与生成建模，我们定义了版面解析、文字-图像对应、字体排印/可读性与字体感知、设计质量评估，以及带隐喻的可控、构图感知生成等任务。我们评估了最优的多模态大语言模型与基于扩散的生成器，发现它们在视觉层级、字体语义、显著性控制与意图传达上存在持续差距；商业模型在高层推理上领先，但作为自动评分器并不敏感，而生成器能很好地渲染文字，却在构图感知合成上表现吃力。大量分析表明，PosterIQ 既是定量基准，也是设计推理的诊断工具，提供可复现、任务特定的指标。我们旨在激发模型的创造力，并将以人为本的设计原则融入生成式视觉-语言系统。
英文原文摘要 1180 字符
We present PosterIQ, a design-driven benchmark for poster understanding and generation, annotated across composition structure, typographic hierarchy, and semantic intent. It includes 7,765 image-annotation instances and 822 generation prompts spanning real, professional, and synthetic cases. To bridge visual design cognition and generative modeling, we define tasks for layout parsing, text-image correspondence, typography/readability and font perception, design quality assessment, and controllable, composition-aware generation with metaphor. We evaluate state-of-the-art MLLMs and diffusion-based generators, finding persistent gaps in visual hierarchy, typographic semantics, saliency control, and intention communication; commercial models lead on high-level reasoning but act as insensitive automatic raters, while generators render text well yet struggle with composition-aware synthesis. Extensive analyses show PosterIQ is both a quantitative benchmark and a diagnostic tool for design reasoning, offering reproducible, task-specific metrics. We aim to catalyze models' creativity and integrate human-centred design principles into generative vision-language systems.