arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2210.08709cs.CLcs.AI
收藏

一个基于正-无标记的文档级关系抽取学习框架

A Unified Positive-Unlabeled Learning Framework for Document-Level Relation Extraction with Different Levels of Labeling

East China Normal University(华东师范大学);Tsinghua University(清华大学)

Ye Wang, Xinxin Liu, Wenxin Hu, Tao Zhang

摘要:

文档级关系抽取(RE)旨在识别跨多个句子的实体之间的关系。大多数以前的方法都集中在完全监督下的文档级RE。然而,在现实世界的情况下,这是昂贵的,很难完全标记在一个文档中的所有关系,因为在文档级RE的实体对的数量与实体的数量呈二次方增长。为了解决常见的不完全标记问题,我们提出了一个统一的正未标记学习框架-移位和平方排名损失正未标记(SSR-PU)学习。我们第一次在文档级RE上使用了正向无标签(PU)学习。考虑到数据集的标记数据可能导致未标记数据的先验移位,我们引入了一种基于训练数据先验移位的PU学习。此外,使用非类得分作为自适应阈值,我们提出了平方排序损失,并证明了其贝叶斯一致性与多标签的排名指标。大量的实验表明,我们的方法实现了约14个F1点相对于以前的基线不完全标记的改进。此外,它在完全监督和极端无标签设置下的性能也优于以前的最先进结果。

英文摘要:

Document-level relation extraction (RE) aims to identify relations between entities across multiple sentences. Most previous methods focused on document-level RE under full supervision. However, in real-world scenario, it is expensive and difficult to completely label all relations in a document because the number of entity pairs in document-level RE grows quadratically with the number of entities. To solve the common incomplete labeling problem, we propose a unified positive-unlabeled learning framework - shift and squared ranking loss positive-unlabeled (SSR-PU) learning. We use positive-unlabeled (PU) learning on document-level RE for the first time. Considering that labeled data of a dataset may lead to prior shift of unlabeled data, we introduce a PU learning under prior shift of training data. Also, using none-class score as an adaptive threshold, we propose squared ranking loss and prove its Bayesian consistency with multi-label ranking metrics. Extensive experiments demonstrate that our method achieves an improvement of about 14 F1 points relative to the previous baseline with incomplete labeling. In addition, it outperforms previous state-of-the-art results under both fully supervised and extremely unlabeled settings as well.

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。

↑