Abstract
We present PharmaShip, a real-world Chinese dataset of scanned pharmaceutical shipping documents designed to stress-test pre-trained text-layout models under noisy OCR and heterogeneous templates. PharmaShip covers three complementary tasks-sequence entity recognition (SER), relation extraction (RE), and reading order prediction (ROP)-and adopts an entity-centric evaluation protocol to minimize confounds across architectures. We benchmark five representative baselines spanning pixel-aware and geometry-aware families (LiLT, LayoutLMv3-base, GeoLayoutLM and their available RORE-enhanced variants), and standardize preprocessing, splits, and optimization. Experiments show that pixels and explicit geometry provide complementary inductive biases, yet neither alone is sufficient: injecting reading-order-oriented regularization consistently improves SER and EL and yields the most robust configuration, while longer positional coverage stabilizes late-page predictions and reduces truncation artifacts. ROP is accurate at the word level but challenging at the segment level, reflecting boundary ambiguity and long-range crossings. PharmaShip thus establishes a controlled, reproducible benchmark for safety-critical document understanding in the pharmaceutical domain and highlights sequence-aware constraints as a transferable bias for structure modeling. We release the dataset at this https URL.
Abstract (translated)
我们介绍了PharmaShip，这是一个包含扫描的中国药品运输文件的真实世界数据集，旨在测试预训练文本布局模型在噪声OCR和异构模板下的性能。PharmaShip涵盖了三个互补的任务：序列实体识别（SER）、关系抽取（RE）以及阅读顺序预测（ROP），并采用以实体为中心的评估协议来最小化不同架构之间的混淆因素。我们对五种具有代表性的基线方法进行了基准测试，这些方法跨越像素感知和几何感知两大类（包括LiLT、LayoutLMv3-base、GeoLayoutLM及其可用的RORE增强版本），并且统一了预处理、数据拆分以及优化标准。 实验结果表明，像素信息与显式几何结构提供了互补性的归纳偏置，但两者单独使用都不足以达到最佳效果：注入以阅读顺序为导向的正则化能够持续改进SER和EL性能，并提供最稳健的配置；而增加位置覆盖范围可以稳定页面后段的预测并减少截断产生的伪影。尽管ROP在单词级别上的准确性较高，但在分段级别上却具有挑战性，这反映了边界模糊以及长距离跨越的问题。 因此，PharmaShip为药品领域中关键安全性的文件理解建立了可控制且可重复的基准测试，并强调了序列感知约束作为结构建模中的可转移偏置。数据集可在以下链接获取：[此链接](this https URL)。
URL
https://arxiv.org/abs/2512.23714
