Coreference Resolution on GAP
Overall F1
- 2019-06-09 — Full Ensemble: Overall F1 90.2
- 2019-08-01 — ProBERT: Overall F1 92.5
| Rank | Model | Overall F1 | Masculine F1 (M) | Feminine F1 (F) | Bias (F/M) | F1 | Paper | Code | Year |
|---|---|---|---|---|---|---|---|---|---|
| 1 | Coref-MTL | 92.72 | 92.65 | 92.45 | 0.99 | – | |||
| 2 | ProBERT | 92.5 | 94.0 | 91.1 | 0.97 | – | Gendered Ambiguous Pronouns Shared Task: Boosting Model Confidence by Evidence Pooling | sattree/gap | 2019 |
| 3 | Maverick_incr | 91.2 | – | – | – | – | Maverick: Efficient and Accurate Coreference Resolution Defying Recent Trends | sapienzanlp/maverick-coref | 2024 |
| 4 | Full Ensemble | 90.2 | 90.9 | 89.5 | 0.98 | – | Gendered Pronoun Resolution using BERT and an extractive question answering formulation | rakeshchada/corefqa | 2019 |
| 5 | PeTra | – | – | – | – | 85.3 | PeTra: A Sparsely Supervised Memory Model for People Tracking | shtoshni92/petra | 2020 |