InfoAgent：基于证据的信息图可追溯生成与修复

中文摘要 AI 辅助

可靠的信息图生成要求事实、符号和视觉关系在渲染和修订过程中保持一致。修正一个元素还需要跟踪其支持证据以及受该变更影响的依赖关系。我们提出 InfoAgent，一个无需训练的证据绑定视觉-符号程序合成框架。其信息图视觉描述（IVD）在类型化依赖图中记录事实载荷、证据来源、执行路径和验证义务。检索到的设计先验指导编译，分层执行将光栅合成与可编辑的符号和绑定对象相结合，同时保留其轨迹。依赖感知修复定位修正、重新检查受影响的依赖，并要求受保护的义务在声明的检查器下保持满足。未解决的义务保持显式。在 IGenBench 上，InfoAgent 达到 93.0 Q-ACC 和 59.0 I-ACC。我们还引入了 InfoGraphicBench-Evidence，其中在 200 个测试请求上的完整检查清单通过率从 Same-IVD Prompt 的 21.5% 增加到初始分层输出的 23.5%，修复后达到 28.5%，使用相同的证据和初始 IVD。在 120 个审计修复案例中，局部修复平均编辑画布的 12.4%，而全局重新生成为 67.3%。

英文摘要

Reliable infographic generation requires facts, symbols, and visual relations to remain consistent through rendering and revision. Correcting one element also requires tracking its supporting evidence and the dependencies affected by the change. We present \textbf{InfoAgent}, a training-free framework for \emph{evidence-bound visual-symbolic program synthesis}. Its Infographic Visual Description (IVD) records factual payloads, evidence provenance, execution routes, and verification obligations in a typed dependency graph. Retrieved design priors guide compilation, and layered execution combines raster synthesis with editable symbolic and binding objects while retaining their traces. Dependency-aware repair localizes corrections, rechecks affected dependencies, and requires protected obligations to remain satisfied under the declared checkers. Unresolved obligations remain explicit. On IGenBench, InfoAgent achieves 93.0 Q-ACC and 59.0 I-ACC. We also introduce InfoGraphicBench-Evidence, where complete-checklist pass rates on 200 test requests increase from 21.5\% for Same-IVD Prompt to 23.5\% for the initial layered output and 28.5\% after repair, using the same evidence and initial IVD. On 120 audited repair cases, localized repair edits 12.4\% of the canvas on average, compared with 67.3\% for global regeneration.

发表机构

• University of Science and Technology of China（中国科学技术大学）
• Meituan（美团）
• Peking University（北京大学）

机构由 AI 辅助整理，请以论文原文为准。