We propose a global entity disambiguation (ED) model based on BERT. To capture global contextual information for ED, our model treats not only words but also entities as input tokens, and solves the task by sequentially resolving mentions to their referent entities and using resolved entities as inputs at each step. We train the model using a large entity-annotated corpus obtained from Wikipedia. We achieve new state-of-the-art results on five standard ED datasets: AIDA-CoNLL, MSNBC, AQUAINT, ACE2004, and WNED-WIKI. The source code and model checkpoint are available at https://github.com/studio-ousia/luke.
翻译:我们提议了一个基于BERT的全球实体脱节模式。为了为ED收集全球背景信息,我们的模式不仅将单词和实体作为投入符号,而且将实体作为投入符号,通过按顺序解决向其参考实体提及的问题,并将已解决的实体作为每一步骤的投入来解决这个问题。我们用从维基百科获得的大型实体附加说明的文具来培训该模式。我们在5个标准的ED数据集(AIDA-CONLL、MSNBC、AQUAINT、ACE2004和WNED-WIKI)上取得了新的最新结果。源代码和模式检查站可在https://github.com/studio-ousia/luke上查阅。