Syllogistic validity prediction on SemEval Task 11 2026 (nested 5-fold cross-val)
94.3Acc+ Z3 Tiebreaker
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| + Z3 TiebreakerStrategy=+ Z3 Tiebreaker2026.04 | 94.3 | 2.85 | 41.88 | |
| Ensemble (pure)Strategy=Ensemble (pure)2026.04 | 93.4 | 3.39 | 39.12 | |
| + Z3 WeightedStrategy=+ Z3 Weighted2026.04 | 93.4 | 3.39 | 39.12 | |
| Top 3 + Z3Strategy=Top 3 + Z32026.04 | 92.2 | 5.08 | 34.11 | |
| Confidence + Z3Strategy=Confidence + Z32026.04 | 91.7 | 6.15 | 31.77 | |
| Z3 OnlyStrategy=Z3 Only2026.04 | 74.7 | 26.28 | 17.39 | |
| + Z3 VetoStrategy=+ Z3 Veto2026.04 | 74.1 | 26.7 | 17.18 |