用嵌入学习打乱文档的正确页序，对15页内效果极佳。
Learning Page Order in Shuffled WOO Releases
- 通过页面嵌入与排序模型重建打乱文档的原始页序。
- 15页文档排序准确率（Kendall's tau）达0.72，短文档超0.95。
- 长文档需专用模型，通用方法易失效，适合信息整理场景。
我们基于5,461份打乱的WOO文档（荷兰信息公开释放文件）研究文档页序恢复问题。这些文档为邮件、法律文本和电子表格等异构内容合并成的PDF，语义顺序信号不可靠。比较了五种方法，包括指针网络、seq2seq Transformer和专用成对排序模型。最优方法在15页以内文档上表现良好，短文档（2-5页）的Kendall's tau达0.95，15页文档为0.72。观察到两个意外失败：seq2seq Transformer在长文档上严重退化（21-25页时tau降至0.014），课程学习比直接训练差39%。消融实验表明，学习的位置编码是导致seq2seq失败的因素之一，但退化现象在所有编码变体中持续存在，说明存在多重交互原因。注意力模式分析显示，短文档与长文档需截然不同的排序策略，解释了课程学习的失败。模型专业化在长文档上带来显著提升（+0.21 tau）。
原文摘要 · Abstract (English)
We investigate document page ordering on 5,461 shuffled WOO documents (Dutch freedom of information releases) using page embeddings. These documents are heterogeneous collections such as emails, legal texts, and spreadsheets compiled into single PDFs, where semantic ordering signals are unreliable. We compare five methods, including pointer networks, seq2seq transformers, and specialized pairwise ranking models. The best performing approach successfully reorders documents up to 15 pages, with Kendall's tau ranging from 0.95 for short documents (2-5 pages) to 0.72 for 15 page documents. We observe two unexpected failures: seq2seq transformers fail to generalize on long documents (Kendall's tau drops from 0.918 on 2-5 pages to 0.014 on 21-25 pages), and curriculum learning underperforms direct training by 39% on long documents. Ablation studies suggest learned positional encodings are one contributing factor to seq2seq failure, though the degradation persists across all encoding variants, indicating multiple interacting causes. Attention pattern analysis reveals that short and long documents require fundamentally different ordering strategies, explaining why curriculum learning fails. Model specialization achieves substantial improvements on longer documents (+0.21 tau).
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。