Global Entity Disambiguation with BERT

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

PDF Abstract NAACL 2022 PDF NAACL 2022 Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Entity Disambiguation ACE2004 confidence-order Micro-F1 91.9 # 2
Entity Disambiguation AIDA-CoNLL confidence-order In-KB Accuracy 95.0 # 1
Entity Disambiguation AQUAINT confidence-order Micro-F1 93.5 # 1
Entity Disambiguation MSNBC confidence-order Micro-F1 96.3 # 1
Entity Disambiguation WNED-CWEB MEP Micro-F1 76.2 # 6
Entity Disambiguation WNED-CWEB confidence-order Micro-F1 78.9 # 2
Entity Disambiguation WNED-WIKI MEP Micro-F1 86.2 # 5
Entity Disambiguation WNED-WIKI confidence-order Micro-F1 89.1 # 2