Entity Linking on AIDA CoNLL03 (test)
82.4Strong F1Kolitsas et al. (2018)
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Kolitsas et al. (2018)Description=Joint modelling of MD and ED2020.03 | 82.4 | 82.8 | 87.3 | |
| BERT+EntitySetting=Setting II, Training=Both BERT and entity classifier trained on Wikipedia with downsampling2020.03 | 79.3 | 81.1 | 87.9 | |
| BERT+EntitySetting=Setting I, Training=Both BERT and entity classifier trained on Wikipedia2020.03 | 74.8 | 76.5 | 78.8 | |
| Kolitsas et al. (2018) indep. baselineDescription=Independent modelling of MD and ED, Mention Detection=Stanford NLP2020.03 | 74.6 | 75 | — | |
| Frozen-BERT+EntitySetting=Setting II, Training=Frozen BERT, entity classifier trained on Wikipedia with downsampling2020.03 | 67.8 | 71.9 | 67.8 | |
| Frozen-BERT+EntitySetting=Setting I, Training=Frozen BERT, entity classifier trained on Wikipedia2020.03 | 64.7 | 68 | 68.6 | |
| BERTTraining=Finetuned on CoNLL03/AIDA, Classifier=Trained from scratch2020.03 | 49.6 | 52.4 | 52.8 |