Opponent priority-ordering prediction on CaSiNo (150-dialogue held-out split)
37.3EMA
70B structured-CoT prompted
Updated 27d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.05 | 37.3 | 62.99 | 70.94 |
70B structured-CoT prompted
| Method | Links | |||
|---|---|---|---|---|
2026.05 | 37.3 | 62.99 | 70.94 |