Commonsense Reasoning on ARC Challenge (val)
91.7AccuracySKILL-MOE
Evaluation Results
| Method | Links | |
|---|---|---|
| SKILL-MOE2025.03 | 91.7 | |
| Mixture-of-Agents2025.03 | 90.1 | |
| Self-ConsistencyAggregation=Best ×52025.03 | 89.3 | |
| ReConcile2025.03 | 89 | |
| Task-Best Model2025.03 | 88.6 |
Loading the SOTA2 catalog…