位置、类型、原因与重要性：面向文本到图像反馈的结构化缺陷定位
Where, What, Why, and Importance: Structured Defect Grounding for Text-to-Image Feedback
AI总结：
提出结构化缺陷定位（SDG）方法，将文本到图像生成中的缺陷诊断建模为结构化集合预测，通过构建SDG-30K数据集和SDG-Eval评估协议，并利用视觉语言模型作为检测器，结合BoxFlow-GRPO将预测的缺陷集合转化为空间奖励以改进扩散模型对齐。
摘要：
尽管文本到图像（T2I）模型生成的图像越来越逼真，但它们仍然存在局部、细微且结构复杂的失败。诊断这些失败需要实例级别的反馈，回答缺陷发生的位置、类型、原因及其对整体图像质量的重要性。虽然最近的密集反馈方法超越了标量监督，但其以热图为中心的表示仍将诊断公式化为像素场回归，这使得定位可变数量的缺陷并将语义原因绑定到单个失败变得困难。为了解决这一表示瓶颈，我们提出了结构化缺陷定位（SDG），通过将每个缺陷建模为（位置、类型、原因、重要性）元组，将T2I诊断转化为结构化集合预测。为了使这一公式可训练和可测量，我们引入了SDG-30K，一个包含30K张图像的数据集，具有跨四个现代T2I生成器的框级标注，以及一个专用的评估协议SDG-Eval。基于这种结构化表示，我们进一步提出了一个诊断到对齐的框架，其中视觉语言模型（VLM）作为SDG检测器，BoxFlow-GRPO将预测的缺陷集合转化为基于框的、重要性加权的空间奖励，用于扩散模型对齐。大量实验表明，我们的SDG检测器在结构化缺陷定位上优于领先的专有VLM，而SDG引导的奖励一致地改善了T2I对齐并支持局部图像细化。这些结果确立了SDG作为诊断、评估和增强现代生成模型的统一实例级接口。
英文摘要：
Despite generating increasingly photorealistic images, text-to-image (T2I) models still exhibit localized, subtle, and structurally complex failures. Diagnosing these failures requires instance-level feedback that answers where a defect occurs, what type it is, why it is defective, and its importance to overall image quality. While recent dense-feedback methods move beyond scalar supervision, their heatmap-centric representations still formulate diagnosis as pixel-field regression, making it difficult to localize variable-cardinality defects and bind semantic reasons to individual failures. To address this representation bottleneck, we propose Structured Defect Grounding (SDG), which casts T2I diagnosis as structured set prediction by modeling each defect as a (location, type, reason, importance) tuple. To make this formulation trainable and measurable, we introduce SDG-30K, a 30K-image dataset with box-grounded annotations across four modern T2I generators, together with a dedicated evaluation protocol, SDG-Eval. Building on this structured representation, we further present a diagnosis-to-alignment framework in which a Vision-Language Model (VLM) serves as the SDG detector, and BoxFlow-GRPO converts predicted defect sets into box-derived, importance-weighted spatial rewards for diffusion model alignment. Extensive experiments show that our SDG detector outperforms leading proprietary VLMs on structured defect grounding, while SDG-guided rewards consistently improve T2I alignment and support localized image refinement. These results establish SDG as a unified, instance-level interface for diagnosing, evaluating, and enhancing modern generative models.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
