Coreference Resolution on WikiCoref (WC) (test)
73.2Average F1ImCoref-CeS_gpt4
Evaluation Results
| Method | Links | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ImCoref-CeS_gpt42025.12 | 73.2 | — | — | — | — | — | — | 83.6 | — | — | 69.8 | — | — | 66.1 | |
| InstructGPT2025.12 | 72.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MEIC-DT2025.12 | 72.6 | — | — | — | — | — | — | 80.2 | — | — | 68.4 | — | — | 69.4 | |
| GPT-3.5-turbo2025.12 | 70.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ImCoref_mes2025.12 | 67.6 | — | — | — | — | — | — | 80.6 | — | — | 66.7 | — | — | 55.6 | |
| Maverick_mes2025.12 | 66.8 | — | — | — | — | — | — | 81.5 | — | — | 65.4 | — | — | 53.5 | |
| Maverick_incr2025.12 | 66.8 | — | — | — | — | — | — | 81.2 | — | — | 65.5 | — | — | 53.7 | |
| Dual-cache2025.12 | 63 | — | — | — | — | — | — | 72.1 | — | — | 62.1 | — | — | 54.7 | |
| LingMess2025.12 | 62.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| longdoc2025.12 | 60.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MTLcomponents=sg+ent2023.09 | 55.6 | — | 80.4 | 60 | 68.7 | 74.5 | 61.8 | 67.5 | 67.8 | 45.3 | 54.4 | 59 | 33 | 42.4 | |
| Joshi et al. (2019)2023.09 | 53 | — | 79.9 | 58.8 | 67.7 | 73.7 | 60.1 | 66.2 | 66.4 | 43.4 | 52.4 | 56.6 | 31.6 | 40.5 | |
| LINGMESS2022.05 | — | 62.6 | — | — | — | — | — | — | — | — | — | — | — | — | |
| longdocTraining=ON2022.08 | — | 61 | — | — | — | — | — | — | — | — | — | — | — | — | |
| longdocTraining=ON, Speaker features (S)=true2022.08 | — | 61.2 | — | — | — | — | — | — | — | — | — | — | — | — | |
| longdocTraining=ON, Speaker features (S)=true, Genre features (G)=true2022.08 | — | 60.9 | — | — | — | — | — | — | — | — | — | — | — | — | |
| longdocTraining=ON + PS 60K, Speaker features (S)=true, Pseudo-singletons (PS)=60K2022.08 | — | 63.8 | — | — | — | — | — | — | — | — | — | — | — | — | |
| longdocTraining=LB02022.08 | — | 47 | — | — | — | — | — | — | — | — | — | — | — | — | |
| longdocTraining=PC2022.08 | — | 52.6 | — | — | — | — | — | — | — | — | — | — | — | — | |
| longdocTraining=Joint, Speaker features (S)=true2022.08 | — | 62.3 | — | — | — | — | — | — | — | — | — | — | — | — | |
| longdocTraining=Joint + PS 60K, Speaker features (S)=true, Pseudo-singletons (PS)=60K2022.08 | — | 64.6 | — | — | — | — | — | — | — | — | — | — | — | — | |
| s2e2022.05 | — | 59.7 | — | — | — | — | — | — | — | — | — | — | — | — |