PosterMELD：面向可控设计多样性且支持可编辑打印就绪输出的多智能体论文到海报生成系统

摘要

科学海报构建将长篇多模态论文压缩为可读、可编辑的画布。现有系统仅对已完成输出进行评分，从而掩盖了请求级失败；直接图像生成无法按元素编辑，而编码智能体工作流成本高昂。PosterMELD 是一种基于模板条件的多智能体流水线：容量感知的槽位在渲染前引导写作，确定性门控与视觉语言模型（VLM）审查将失败路由至有界修复。每个被接受的请求均导出可编辑的 PowerPoint（PPTX）与便携式网络图形（PNG）制品；显式设计控制可生成同一论文的多个变体。在 621 篇论文中，打印就绪率（PRR）统计通过几何、可读性、资产完整性及明显事实错误检查的请求，原生可编辑性则单独报告。一个冻结的 VLM 为打印就绪输出分配条件性工艺-和谐-表现力（CHE）分数。PosterMELD 达到 81.3% 的 PRR，为 P2P 的 3.4 倍、PosterGen 的 5.2 倍，并且在具备多个打印就绪输出的生成方法中取得最高的条件性 CHE。原生可编辑性与显式设计控制在每次请求平均成本 0.38 美元下得以保留，仅为 Codex+Skill 的 3.5%。代码与资源见 https://github.com/Shannon4Science/PosterMELD。

English

Scientific poster construction compresses a long multimodal paper into a readable, editable canvas. Existing systems hide request-level failures by scoring only completed outputs; direct image generation is not element-editable, while coding-agent workflows are costly. PosterMELD is a template-conditioned multi-agent pipeline: capacity-aware slots guide writing before rendering, and deterministic gates plus vision-language model (VLM) review route failures to bounded repair. Each accepted request exports editable PowerPoint (PPTX) and Portable Network Graphics (PNG) artifacts; explicit design controls yield same-paper variants. Across 621 papers, Print-Ready Rate (PRR) counts requests passing geometric, readability, asset-integrity, and obvious-factual-error checks, with native editability reported separately. A frozen VLM assigns conditional Craftsmanship-Harmony-Expressiveness (CHE) scores to print-ready outputs. PosterMELD attains 81.3% PRR, 3.4 times P2P's rate and 5.2 times PosterGen's, and the highest conditional CHE among generated methods with multiple print-ready outputs. Native editability and explicit design controls are retained at a mean cost of USD 0.38 per request, 3.5% of Codex+Skill's. Code and resources are available at https://github.com/Shannon4Science/PosterMELD.