A Unified MRC Framework for Named Entity Recognition

The task of named entity recognition (NER) is normally divided into nested NER and flat NER depending on whether named entities are nested or not. Models are usually separately developed for the two tasks, since sequence labeling models, the most widely used backbone for flat NER, are only able to assign a single label to a particular token, which is unsuitable for nested NER where a token may be assigned several labels... (read more)

PDF Abstract ACL 2020 PDF ACL 2020 Abstract

Results from the Paper


 Ranked #1 on Named Entity Recognition on ACE 2005 (using extra training data)

     Get a GitHub badge
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK USES EXTRA
TRAINING DATA
RESULT BENCHMARK
Nested Mention Recognition ACE 2004 BERT-MRC F1 85.98 # 1
Named Entity Recognition ACE 2005 BERT-MRC F1 86.88 # 1
Named Entity Recognition CoNLL 2003 (English) BERT-MRC F1 93.04 # 13
Nested Named Entity Recognition GENIA BERT-MRC F1 83.75 # 1
Chinese Named Entity Recognition MSRA BERT-MRC F1 95.75 # 3
Chinese Named Entity Recognition OntoNotes 4 BERT-MRC F1 82.11 # 2
Named Entity Recognition Ontonotes v5 (English) BERT-MRC F1 91.11 # 3

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet