An Improved Baseline for Sentence-level Relation Extraction
An Improved Baseline for Sentence-level Relation Extraction
复制标题
DOI：
--
复制
发表时间：
2021-02
期刊：
影响因子：
--
通讯作者：
Wenxuan Zhou;Muhao Chen
中科院分区：
文献类型：
--
作者：
Wenxuan Zhou;Muhao Chen
句子级关系抽取（RE）旨在识别句子中两个实体之间的关系。人们已经为此付出了许多努力，而最好的方法仍然远远不够完美。在本文中，我们重新审视两个问题，影响现有的RE模型的性能，即实体表示和嘈杂或定义不清的标签。我们改进的RE基线，结合实体表示与类型化标记，实现了74.6%的F1 TACRED，显着优于以前的SOTA方法。此外，所提出的新基线在改进的Re-TACRED数据集上实现了91.1%的F1，表明预训练语言模型（PLM）在此任务上实现了高性能。我们向社区发布代码以供将来研究。
Sentence-level relation extraction (RE) aims at identifying the relationship between two entities in a sentence. Many efforts have been devoted to this problem, while the best performing methods are still far from perfect. In this paper, we revisit two problems that affect the performance of existing RE models, namely entity representation and noisy or ill-defined labels. Our improved RE baseline, incorporated with entity representations with typed markers, achieves an F1 of 74.6% on TACRED, significantly outperforms previous SOTA methods. Furthermore, the presented new baseline achieves an F1 of 91.1% on the refined Re-TACRED dataset, demonstrating that the pretrained language models (PLMs) achieve high performance on this task. We release our code to the community for future research.