Named Entity Recognition on CoNLL-2003 German 2006 Revised (test)
92.23F1 ScoreTransformer-Linear (+DEV)
Evaluation results
| Method | Links | |
|---|---|---|
| Transformer-Linear (+DEV)Approach=Fine-tuning (Ablations), Document features=yes (+enforce), Training=Includes dev data2020.11 | 92.23 | |
| Transformer-LinearApproach=Fine-tuning, Document features=yes2020.11 | 91.55 | |
| Transformer-LinearApproach=Fine-tuning (Ablations), Document features=yes (+enforce)2020.11 | 91.33 | |
| Yu et al. (2020)Document features=yes2020.11 | 90.3 | |
| Transformer-LinearApproach=Fine-tuning, Document features=no2020.11 | 90.04 | |
| LSTM-CRF (all layer mean)Approach=Feature-based, Document features=yes2020.11 | 89.88 | |
| Akbik et al. (2019b)Document features=pooling2020.11 | 88.27 | |
| LSTM-CRF (all layer mean)Approach=Feature-based, Document features=no2020.11 | 87.35 |