FLERT XLM-R

Reported on 5 benchmarks across 1 task · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing5 results

Named Entity Recognition (NER)onCoNLL 2003 (German) Revised
F1· 2020-11-13
92.23
SOTA
FLERT: Document-Level Features for Named Entity Recognition arXiv:2011.06993
Named Entity Recognition (NER)onCoNLL 2003 (German)
F1· 2020-11-13
88.34
best: 88.38 (ACE + document-context)
FLERT: Document-Level Features for Named Entity Recognition arXiv:2011.06993
Named Entity Recognition (NER)onCoNLL 2003 (English)
F1· uses extra data· 2020-11-13
94.09
best: 94.6 (ACE + document-context)
FLERT: Document-Level Features for Named Entity Recognition arXiv:2011.06993
Named Entity Recognition (NER)onCoNLL 2002 (Spanish)
F1· 2020-11-13
90.14
best: 95.9 (ACE + document-context)
FLERT: Document-Level Features for Named Entity Recognition arXiv:2011.06993
Named Entity Recognition (NER)onCoNLL 2002 (Dutch)
F1· 2020-11-13
95.21
best: 95.7 (ACE + document-context)
FLERT: Document-Level Features for Named Entity Recognition arXiv:2011.06993