List your company
ResearchFormerly WizWand

Named Entity Recognition on Spanish (test)

86.3F1 ScoreMonolingual
60.92467.51274.180.688Aug 31, 2019Oct 9, 2019Nov 18, 2019Dec 28, 2019Feb 6, 2020Mar 17, 2020Apr 26, 2020
Updated 8mo ago

Evaluation results

MethodLinks
Monolingualtraining_data=annotated target language2019.0886.3
Xie et al. (2018)Emb. type=glove, Emb. dim=3002019.1186.26
Lample et al. (2016)Emb. type=skip-ngram, Emb. dim=642019.1185.75
Source-MonoEmb. type=fasttext, Emb. dim=3002019.1184.71
Ours-simSource setting=Multi-source, Teacher Fusion=weighting teacher models with learned language similarities (Eq. 11)2020.0478
Ours-avgSource setting=Multi-source, Teacher Fusion=averaging teacher models (Eq. 7)2020.0477.75
Moon et al. (2019)Source setting=Multi-source, Note=freezing the bottom three layers of BERT_BASE2020.0476.53
TMPBackbone=BiLSTM + CRF with self-attention2019.0873.5
Chen et al. (2019)Source setting=Multi-source2020.0473.5
BWETtranslation_type=word-by-word, lexicon=bilingual lexicon induced using monolingual corpora2019.0872.4
Rahimi et al. (2019)Source setting=Multi-source2020.0471.8
Co-decoding2019.0865.1
fast-alignBackbone=BiLSTM + CRF with self-attention, alignment_algorithm=unsupervised expectation maximization2019.0865
Polyglot-NER2019.0863
Täckström (2012)Source setting=Multi-source2020.0461.9