编码智能体在视觉生成中能走多远？
原标题：How Far Can Coding Agents Go in Visual Generation?
AI 摘要
该研究探索通用编码智能体通过可执行程序进行视觉生成的能力，提出AgentDrawBench基准（200个提示）评估零样本绘图。结果显示，编码智能体在结构保真度和精确视觉约束上表现优异（如GPT-6-Astra配合Canvas 2D达85.8%结构保真度），但在感知丰富度上不及原生图像生成器。研究进一步提出视觉代码可作为外部视觉思维语言，在视觉问答中代码重建保留大量信息，部分领域准确率超过原始图像，并发现智能体可能维护跨绘图语言的共享内部场景表示。
正文节选
How Far Can Coding Agents Go in Visual Generation? Abstract Modern image generation largely relies on specialized generative models that leave the process of image construction implicit in learned parameters. We explore an alternative: general-purpose coding agents that construct images through executable programs. Our study of zero-shot drawing on AgentDrawBench, a 200-prompt benchmark, reveals a distinctive capability profile. Agents excel at preserving structure and satisfying precise visual