Global Entity Disambiguation with BERT
2019-09-01NAACL 2022Code Available1· sign in to hype
Ikuya Yamada, Koki Washio, Hiroyuki Shindo, Yuji Matsumoto
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/studio-ousia/lukeOfficialIn paperpytorch★ 727
Abstract
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.
Tasks
Benchmark Results
| Dataset | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| ACE2004 | confidence-order | Micro-F1 | 91.9 | — | Unverified |
| AIDA-CoNLL | confidence-order | In-KB Accuracy | 95 | — | Unverified |
| AQUAINT | confidence-order | Micro-F1 | 93.5 | — | Unverified |
| MSNBC | confidence-order | Micro-F1 | 96.3 | — | Unverified |
| WNED-CWEB | confidence-order | Micro-F1 | 78.9 | — | Unverified |
| WNED-CWEB | MEP | Micro-F1 | 76.2 | — | Unverified |
| WNED-WIKI | MEP | Micro-F1 | 86.2 | — | Unverified |
| WNED-WIKI | confidence-order | Micro-F1 | 89.1 | — | Unverified |