Event extraction, the technology that aims to automatically get the structural information from documents, has attracted more and more attention in many fields. Most existing works discuss this issue with the token-level multi-label classification framework by distinguishing the tokens as different roles while ignoring the writing styles of documents. The writing style is a special way of content organizing for documents and it is relative fixed in documents with a special field (e.g. financial, medical documents, etc.). We argue that the writing style contains important clues for judging the roles for tokens and the ignorance of such patterns might lead to the performance degradation for the existing works. To this end, we model the writing style in documents as a distribution of argument roles, i.e., Role-Rank Distribution, and propose an event extraction model with the Role-Rank Distribution based Supervision Mechanism to capture this pattern through the supervised training process of an event extraction task. We compare our model with state-of-the-art methods on several real-world datasets. The empirical results show that our approach outperforms other alternatives with the captured patterns. This verifies the writing style contains valuable information that could improve the performance of the event extraction task.
RDU: A Region-based Approach to Form-style Document Understanding
National University of Singapore(新加坡国立大学);Estates Pte Ltd(6Estates私人有限公司);Sichuan University(四川大学)
关键信息提取(Key Information Extraction,KIE)是从表单格式的文档(如发票)中提取结构化信息(如键值对)的技术,是实现文档理解智能化的重要一步。以往的KIE算法一般采用序列标记的方法,难以处理非扁平化序列,尤其是表格-文本混合文档。这些方法还存在为每种类型的文档预先定义固定的标签集的麻烦,以及标签不平衡的问题。在这项工作中,我们假设光学字符识别(OCR)已被应用到输入文件,并重新制定的KIE任务作为一个区域预测问题,在二维(2D)空间给定的目标字段。在这种新的设置,我们开发了一个新的KIE模型命名为基于区域的文档理解(RDU),作为输入的文本内容和相应的坐标的文件,并试图预测的结果,通过本地化的边界框样区域。我们的RDU首先应用一个布局感知BERT配备了一个软布局注意掩蔽和偏见机制,将布局信息的表示。然后,候选区域的列表通过区域提议模块从表示中生成,该区域提议模块受到广泛应用于目标检测的计算机视觉模型的启发。最后,区域分类模块和区域选择模块分别用于判断所提出的区域是否有效,并从所有区域中选择概率最大的区域。在四种表单风格的文档上的实验表明,我们提出的方法可以取得令人印象深刻的结果。此外,我们的RDU模型可以无缝地使用不同的文档类型进行训练,这对低资源文档特别有用。
Dependent Gated Reading for Cloze-Style Question Answering
Oregon State University(俄勒冈州立大学)
我们提出了一种新的深度学习架构来解决完形填空式的问答任务。现有的方法采用的阅读机制,没有充分利用文档和查询之间的相互依赖性。在本文中,我们提出了一种新的\emph{dependent gated reading}双向GRU网络(DGR),有效地建模之间的关系,在编码和决策的文档和查询。我们的评估表明,DGR在著名的机器理解基准测试中获得了极具竞争力的性能,如儿童图书测试(CBT-NE和CBT-CN)和Who DiD What(WDW,严格和放松)。最后,我们广泛地分析和验证我们的模型消融和注意力的研究。
The Effect of Different Writing Tasks on Linguistic Style: A Case Study of the ROC Story Cloze Task
University of Washington(华盛顿大学);Allen Institute for Artificial Intelligence(艾伦人工智能研究所)
一个作家的风格不仅取决于个人特质,还取决于她的意图和精神状态。在本文中,我们展示了如何相同的写作任务的变体可以导致可测量的写作风格的差异。我们提出了一个基于故事完形填空任务的案例研究(Mostafazadeh et al.,2016 a),其中注释者被分配了具有不同约束的类似写作任务:(1)编写整个故事,(2)为给定的故事上下文添加故事结局,以及(3)为故事添加不连贯的结局。我们表明,一个简单的线性分类通知文体特征是能够成功区分这三种情况下,甚至不看的故事背景。此外,将我们的文体特征与语言模型预测相结合,达到了故事完形填空的最佳表现。我们的研究结果表明,不同的任务框架可以显着影响人们的写作方式。
Better Call Claude: Can LLMs Detect Changes of Writing Style?
Center for Artificial Intelligence, Technical University of Applied Sciences Würzburg-Schweinfurt(人工智能中心,应用技术大学Würzburg-Schweinfurt);Humanities Lab, Faculaty of Arts, Radboud University(人文实验室,艺术学院,拉德博德大学);LEYA Lab, School of Computer Science, Physics and Technology, HSE University(LEYA实验室,计算机科学、物理与技术学院,俄罗斯高等经济大学)
Examining Scientific Writing Styles from the Perspective of Linguistic Complexity
Nanjing University of Science and Technology(南京理工大学);Indiana University(印第安纳大学);Nanjing University(南京大学);Wuhan University(武汉大学);University of Illinois(伊利诺伊大学)
How to Write High-quality News on Social Network? Predicting News Quality by Mining Writing Style
Key Laboratory of Intelligent Information Processing Center for Advanced Computing Research, Institute of Computing Technology, CAS(中国科学院计算技术研究所);University of Chinese Academy of Sciences(中国科学院大学);National Tsing Hua University(国立清华大学)
Overview of PAN 2026: Voight-Kampff Generative AI Detection, Text Watermarking, Multi-Author Writing Style Analysis, Generative Plagiarism Detection, and Reasoning Trajectory Detection
Friedrich Schiller University Jena(耶纳弗里德里希·施勒格尔大学);University of Applied Sciences BFI(应用科学大学BFI);Mohamed bin Zayed University of Artificial Intelligence(马尔代夫 bin Zayed 人工智能大学);University of Kassel(卡塞尔大学);INSAIT, Sofia University ``St. Kliment Ohridski''(INSAIT,索菲亚大学『圣克莱门特·奥赫里迪斯』);University of Innsbruck(因斯布鲁克大学)
PAN 2026旨在通过客观评估推进计算风格学和文本取证,涵盖生成式AI检测、文本水印、多作者写作分析、抄袭检测及推理轨迹检测等五个新旧任务。
Stylometry Analysis of Multi-authored Documents for Authorship and Author Style Change Detection
Abasyn University Islamabad Campus(阿巴辛大学伊斯兰堡校区);University of Engineering and Technology, Peshawar(白沙瓦工程技术大学);National University of Science and Technology(国立科技大学);University of Western Australia(西澳大学);Munster Technological University(芒斯特理工大学)
The Importance of Suppressing Domain Style in Authorship Analysis
Technical University of Munich(慕尼黑工业大学);Humboldt-Universität zu Berlin(柏林洪堡大学);University of Oxford(牛津大学);Martin-Luther-Universität Halle-Wittenberg(马丁·路德·哈勒-维滕贝格大学);University of the Aegean(爱琴海大学);Bauhaus-Universität Weimar(魏玛包豪斯大学);Leipzig University(莱比锡大学)
Personalized Text Generation with Contrastive Activation Steering
Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院);Northeastern University(东北大学);University of Science and Technology of China(中国科学技术大学);National University of Singapore(新加坡国立大学)