CrossWeigh + Pooled Flair

Reported on 3 benchmarks across 1 task · 1 paper · 2 SOTA

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

Natural Language Processing3 results

Named Entity Recognition (NER)onWNUT 2017
F1· 2019-09-03
50.03
best: 60.45 (CL-KL)
SOTA
CrossWeigh: Training Named Entity Tagger from Imperfect Annotations arXiv:1909.01441
Named Entity Recognition (NER)onCoNLL++
F1· uses extra data· 2019-09-03
94.28
best: 96.12 (LUKE + SubRegWeigh (K-means))
SOTA
CrossWeigh: Training Named Entity Tagger from Imperfect Annotations arXiv:1909.01441
Named Entity Recognition (NER)onCoNLL 2003 (English)
F1· 2019-09-03
93.43
best: 94.6 (ACE + document-context)
CrossWeigh: Training Named Entity Tagger from Imperfect Annotations arXiv:1909.01441