arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2004.12006cs.CL
收藏

基于文本百科知识的语境化表示

Contextualized Representations Using Textual Encyclopedic Knowledge

Google Research(谷歌研究院);University of Washington(华盛顿大学)

Mandar Joshi, Kenton Lee, Yi Luan, Kristina Toutanova

摘要:

我们提出了一种方法来表示输入文本的上下文,他们与动态检索的文本百科全书背景知识从多个文件。我们将我们的方法应用于阅读理解任务,通过编码问题和段落以及关于他们提到的实体的背景句子。我们表明,从文本中整合背景知识对于专注于事实推理的任务是有效的,并允许直接重用功能强大的预训练BERT风格的编码器。此外,知识集成可以进一步提高与适当的预训练,通过一个自我监督的掩蔽语言模型目标在背景增强的输入文本中的单词。在TriviaQA上,我们的方法比不动态整合背景知识的RoBERTA模型获得了1.6到3.1 F1的改进。在MRQA(一个由各种QA数据集组成的大型集合)上,我们看到了域内的一致收益以及BioASQ(2.1至4.2 F1),TextbookQA(1.6至2.0 F1)和DuoRC(1.1至2.0 F1)的域外大幅改善。

英文摘要:

We present a method to represent input texts by contextualizing them jointly with dynamically retrieved textual encyclopedic background knowledge from multiple documents. We apply our method to reading comprehension tasks by encoding questions and passages together with background sentences about the entities they mention. We show that integrating background knowledge from text is effective for tasks focusing on factual reasoning and allows direct reuse of powerful pretrained BERT-style encoders. Moreover, knowledge integration can be further improved with suitable pretraining via a self-supervised masked language model objective over words in background-augmented input text. On TriviaQA, our approach obtains improvements of 1.6 to 3.1 F1 over comparable RoBERTa models which do not integrate background knowledge dynamically. On MRQA, a large collection of diverse QA datasets, we see consistent gains in-domain along with large improvements out-of-domain on BioASQ (2.1 to 4.2 F1), TextbookQA (1.6 to 2.0 F1), and DuoRC (1.1 to 2.0 F1).

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。

↑