Causal Reasoning on Cladder AceReason (Complete)
81.2AccuracyModel-first Greedy
Evaluation Results
| Method | Links | |
|---|---|---|
| Model-first Greedyk=5, Summarizer=AceReason2026.05 | 81.2 | |
| Input-allk=5, Summarizer=AceReason2026.05 | 80.1 | |
| Best-modelk=5, Summarizer=AceReason2026.05 | 78.6 | |
| Top-accuracyk=5, Summarizer=AceReason2026.05 | 77.7 | |
| Truth-prediction Greedyk=5, Summarizer=AceReason2026.05 | 76.2 | |
| GPT5.2-judgek=5, Summarizer=AceReason2026.05 | 76 | |
| MoAk=5, Summarizer=AceReason2026.05 | 75.7 | |
| Oracle-surrogate Greedyk=5, Summarizer=AceReason2026.05 | 75.2 | |
| Conditioned-diversityk=5, Summarizer=AceReason2026.05 | 74.2 | |
| Aya-judgek=5, Summarizer=AceReason2026.05 | 72 |