用视觉与结构信息恢复缺失的PDF表单字段，准确率超93%。
AcroMELD: Recovering Interactive PDF Forms with Structure-Aware Graph Set Transformers
- 融合视觉与PDF结构线索，通过图集合变换器联合推理字段位置。
- 内部测试微F1达0.9344，外部单次样本测试仍达0.8477。
- 适合需要自动化表单重建的文档处理系统使用。
交互式PDF表单字段常因文档外观相似却无实际控件而缺失，导致用户无法直接输入数据。检测缺失字段困难，因字段可能由多重重叠线索指示，且数字版PDF仅提供不完整的绘图结构，密集页面可能包含数百个字段。本文提出AcroMELD（AcroForm多源证据链接解码器），一个39.4M参数的检测模型，结合高分辨率视觉Transformer与无标签的PDF原始结构。其896个查询包括384个视觉提案、384个结构引导提案和128个学习恢复查询。四层图-集合层在几何偏置的稀疏邻域内交换信息，并跨注意力关注PDF结构。学习到的同字段关系连接共指候选，定位质量头基于包含感知重叠进行训练。定义了哈希绑定评估协议，包含独立开发、校准、内部测试及隔离外部保留集。密封单种子候选在内部测试中达到0.9344的原生包含微F1，外部单次样本测试为0.8477（95%置信区间[0.8339, 0.8605]）。相比历史参考FFGBT-v8提升0.0186绝对F1。但在更严格的外部适配器下，性能降至IoU-0.5 F1 0.7786和COCO mAP 0.2900，低于本地评估的CommonForms-L参考；签名类在选定阈值下无预测。结果支持注册操作门限，同时暴露领域泛化与罕见类的局限性。
原文摘要 · Abstract (English)
Interactive PDF form fields are often absent from documents that visually resemble forms, leaving users unable to enter data without printing or external editing tools. Detecting the missing widgets is difficult because a field may be indicated by several overlapping cues, born-digital PDFs expose useful but incomplete drawing structure, and dense pages can contain hundreds of fields. We introduce AcroMELD (AcroForm Multi-source Evidence Linking Decoder), a 39.4M-parameter detector that combines a high-resolution visual transformer with label-free PDF primitives. Its 896-query set comprises 384 visual proposals, 384 structure-seeded proposals, and 128 learned recovery queries. Four graph-set layers exchange information over geometry-biased sparse neighborhoods and cross-attend to PDF structure. A learned same-field relation links co-referent candidates, while a localization-quality head is trained on the containment-aware overlap used by the downstream recovery decision. We define a hash-bound evaluation protocol with disjoint development, calibration, internal-test, and quarantined external-holdout roles. The sealed, single-seed candidate reaches native containment micro-$F_1$ 0.9344 on the internal test and 0.8477 on the one-shot external holdout (95% PDF-cluster bootstrap interval [0.8339, 0.8605]). This passes the registered historical FFGBT-v8 reference by 0.0186 absolute $F_1$. Under the stricter external adapter, however, performance is 0.7786 IoU-$0.5$ $F_1$ and 0.2900 COCO mAP, below a locally evaluated CommonForms-L reference; the signature class receives no prediction at the selected threshold. Thus the result supports the registered operational gate while exposing substantial domain and rare-class limitations.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。