arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1908.10993cs.CLcs.AIcs.DL
收藏

arXiv.org上的科学陈述分类

Scientific Statement Classification over arXiv.org

FAU Erlangen-Nuremberg(埃尔朗根-纽伦堡大学);National Institute of Standards and Technology(美国国家标准与技术研究院)

Deyan Ginev, Bruce R. Miller

AI总结:

该研究提出科学陈述分类新任务,发布基于arXiv预印本的大规模标注数据集,确定13类划分方案,验证BiLSTM模型可达0.91的F1峰值,还提出数学公式词位序列化方法并指出多模态训练的提升潜力,为科学语篇建模提供方向。

摘要:

我们提出了一项新的科学陈述分类任务,并发布了一个用于监督学习的大规模数据集。该数据集源自arXiv.org预印本文章合集的机器可读表示形式。我们探究了50个作者标注的类别,并通过实证确定了任务设计方案:将1050万个带标注的段落划分为13个类别。我们证明该任务设置与现有最优技术的已知成功率相匹配,通过BiLSTM编码器-解码器模型可达到0.91的峰值F1值。此外,我们提出了一种针对数学公式的词位序列化方法,并发现上下文感知模型在同时使用符号模态训练时性能可得到提升。最后,我们讨论了数据与任务设计的局限性,并概述了构建更复杂的科学语篇模型(超越孤立陈述)的潜在方向。

英文摘要:

We introduce a new classification task for scientific statements and release a large-scale dataset for supervised learning. Our resource is derived from a machine-readable representation of the arXiv.org collection of preprint articles. We explore fifty author-annotated categories and empirically motivate a task design of grouping 10.5 million annotated paragraphs into thirteen classes. We demonstrate that the task setup aligns with known success rates from the state of the art, peaking at a 0.91 F1-score via a BiLSTM encoder-decoder model. Additionally, we introduce a lexeme serialization for mathematical formulas, and observe that context-aware models could improve when also trained on the symbolic modality. Finally, we discuss the limitations of both data and task design, and outline potential directions towards increasingly complex models of scientific discourse, beyond isolated statements.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑