Named Entity Recognition on document-level 60/20/20 fold (test)
3,866SelectionsUnion (any model)
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Union (any model)candidate verification strategy=Union2026.05 | 3,866 | 1,742 | 54.9 | 100 | 88.3 | |
| OpenAI ∪ Claudecandidate verification strategy=Union of OpenAI and Claude2026.05 | 2,743 | 743 | 72.9 | 94.2 | 83.2 | |
| Agree ≥ 2agreement threshold=22026.05 | 2,587 | 797 | 69.2 | 84.3 | 74.5 | |
| BioConCal P95P95 operating point=true2026.05 | 1,340 | 82 | 93.9 | 59.2 | 52.3 | |
| Agree ≥ 4agreement threshold=42026.05 | 1,275 | 260 | 79.6 | 47.8 | 42.2 | |
| Agree ≥ 6agreement threshold=62026.05 | 715 | 93 | 87 | 29.3 | 25.9 | |
| Unanimous (k=8)agreement threshold=82026.05 | 293 | 14 | 95.2 | 13.1 | 11.6 |