Commonsense Reasoning on CommonQA (Accuracy %)
85.18AccuracySC
Evaluation Results
| Method | Links | |
|---|---|---|
| SCBase Model=Qwen2.5-32b-instruct2026.04 | 85.18 | |
| HCP-MADBase Model=Qwen2.5-32b-instruct + Llama-3.1-70b-instruct2026.04 | 84.85 | |
| SRBase Model=Qwen2.5-32b-instruct2026.04 | 84.6 | |
| CoTBase Model=Qwen2.5-32b-instruct2026.04 | 84.44 | |
| MADBase Model=Qwen2.5-32b-instruct2026.04 | 84.36 | |
| DOWNBase Model=Qwen2.5-32b-instruct + Llama-3.1-70b-instruct2026.04 | 84.11 | |
| Heter-MADBase Model=Qwen2.5-32b-instruct + Llama-3.1-70b-instruct2026.04 | 83.52 | |
| MARSBase Model=Qwen2.5-32b-instruct + Llama-3.1-70b-instruct2026.04 | 79.85 | |
| MADBase Model=Llama-3.1-70b-instruct2026.04 | 78.3 | |
| SCBase Model=Llama-3.1-70b-instruct2026.04 | 76.49 | |
| SRBase Model=Llama-3.1-70b-instruct2026.04 | 76 | |
| CoTBase Model=Llama-3.1-70b-instruct2026.04 | 75.27 |