Logical Reasoning on Logic
68.07AccuracyEpiCoDe
Evaluation Results
| Method | Links | |
|---|---|---|
| EpiCoDeBackbone=Qwen2-7B-Instruct2025.06 | 68.07 | |
| MEBackbone=Qwen2-7B-Instruct2025.06 | 67.6 | |
| CDBackbone=Qwen2-7B-Instruct2025.06 | 67.43 | |
| FinetuneBackbone=Qwen2-7B-Instruct2025.06 | 66.67 | |
| EpiCoDeBackbone=Deepseek-7B-Chat2025.06 | 59.05 | |
| MEBackbone=Deepseek-7B-Chat2025.06 | 58.89 | |
| CDBackbone=Deepseek-7B-Chat2025.06 | 58.46 | |
| EpiCoDeBackbone=Llama-3.2-3B-Instruct2025.06 | 57.48 | |
| FinetuneBackbone=Deepseek-7B-Chat2025.06 | 57.22 | |
| CDBackbone=Llama-3.2-3B-Instruct2025.06 | 56.62 | |
| MEBackbone=Llama-3.2-3B-Instruct2025.06 | 55.11 | |
| EpiCoDeBackbone=Qwen2-1.5B-Instruct2025.06 | 53.87 | |
| MEBackbone=Qwen2-1.5B-Instruct2025.06 | 53.63 | |
| FinetuneBackbone=Llama-3.2-3B-Instruct2025.06 | 53.45 | |
| CDBackbone=Qwen2-1.5B-Instruct2025.06 | 53.42 | |
| FinetuneBackbone=Qwen2-1.5B-Instruct2025.06 | 52.68 |