arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.10344cs.CL
收藏

利用知识改进生物医学预训练语言模型

Improving Biomedical Pretrained Language Models with Knowledge

Tsinghua University(清华大学);Alibaba Group(阿里巴巴集团)

Zheng Yuan, Yijia Liu, Chuanqi Tan, Songfang Huang, Fei Huang

摘要:

预训练的语言模型已经在许多自然语言处理任务中显示出成功。许多作品探索将知识合并到语言模型中。在生物医学领域,专家们已经花了几十年的努力来建立大规模的知识库。例如,统一医学语言系统(UMLS)包含数百万个实体及其同义词,并定义了实体之间的数百种关系。利用这些知识可以使各种下游任务受益,例如命名实体识别和关系抽取。为此,我们提出了KeBioLM,生物医学预训练的语言模型,明确利用知识的UMLS知识库。具体来说,我们从PubMed摘要中提取实体,并将其链接到UMLS。然后,我们训练知识感知的语言模型,首先应用纯文本编码层来学习实体表示,并应用文本-实体融合编码来聚合实体表示。此外,我们增加了实体检测和实体连接两个训练目标。在BLURB基准测试中进行的命名实体识别和关系抽取实验表明了该方法的有效性。对收集的探测数据集的进一步分析表明,我们的模型具有更好的建模能力的医学知识。

英文摘要:

Pretrained language models have shown success in many natural language processing tasks. Many works explore incorporating knowledge into language models. In the biomedical domain, experts have taken decades of effort on building large-scale knowledge bases. For example, the Unified Medical Language System (UMLS) contains millions of entities with their synonyms and defines hundreds of relations among entities. Leveraging this knowledge can benefit a variety of downstream tasks such as named entity recognition and relation extraction. To this end, we propose KeBioLM, a biomedical pretrained language model that explicitly leverages knowledge from the UMLS knowledge bases. Specifically, we extract entities from PubMed abstracts and link them to UMLS. We then train a knowledge-aware language model that firstly applies a text-only encoding layer to learn entity representation and applies a text-entity fusion encoding to aggregate entity representation. Besides, we add two training objectives as entity detection and entity linking. Experiments on the named entity recognition and relation extraction from the BLURB benchmark demonstrate the effectiveness of our approach. Further analysis on a collected probing dataset shows that our model has better ability to model medical knowledge.

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。

↑