Multi-word Expression Identification on PARSEME EN 1.1 (test)
45.4Token F1-Scoreneural CRF tagger
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| neural CRF taggerEmbeddings=BERT + GloVe, POS/lemma constraints=Gold, Evaluation=Zero-shot2020.04 | 45.4 | 36.1 | 45.5 | 40.3 | 40.2 | 52 | |
| neural CRF taggerEmbeddings=BERT + GloVe, POS/lemma constraints=None, Evaluation=Zero-shot2020.04 | 45.4 | 36.2 | 45.3 | 40.3 | 40.4 | 51.8 | |
| BERTPOS/lemma constraints=Gold, zero-shot=true2020.04 | 45.4 | 36.1 | 45.5 | 40.3 | 40.2 | 52 | |
| BERTPOS/lemma constraints=None, zero-shot=true2020.04 | 45.4 | 36.2 | 45.3 | 40.3 | 40.4 | 51.8 | |
| neural CRF taggerEmbeddings=BERT + GloVe, POS/lemma constraints=Predicted, Evaluation=Zero-shot2020.04 | 43.4 | 34.1 | 45.9 | 39.2 | 37.1 | 52.2 | |
| BERTPOS/lemma constraints=Predicted, zero-shot=true2020.04 | 43.4 | 34.1 | 45.9 | 39.2 | 37.1 | 52.2 | |
| Rohanian et al.2020.04 | 41.9 | — | — | — | — | — | |
| Taslimipoor et al.2020.04 | 40.2 | — | — | — | — | — | |
| Rohanian+supervised=true2020.04 | 40.2 | — | — | 41.9 | — | — | |
| Nerima et al.2020.04 | 34.4 | 33.8 | 32.7 | 33.3 | 37.3 | 31.8 | |
| Kirilin+supervised=true2020.04 | 34.4 | 33.8 | 32.7 | 33.3 | 37.3 | 31.8 | |
| Kirilin et al.2020.04 | 34.1 | — | — | 36 | — | — | |
| Taslimipoor+supervised=true2020.04 | — | — | — | 36 | — | — |