Code Reasoning on CruxEval Output
51ScoreDataFlow-Code-10K
Evaluation Results
| Method | Links | |
|---|---|---|
| DataFlow-Code-10KBase Model=Qwen2.5-14B-Instruct2025.12 | 51 | |
| DataFlow-Code-1KBase Model=Qwen2.5-14B-Instruct2025.12 | 50.9 | |
| DataFlow-Code-5KBase Model=Qwen2.5-14B-Instruct2025.12 | 50.6 | |
| Self-OSSBase Model=Qwen2.5-14B-Instruct2025.12 | 50.1 | |
| Code Alpaca-1KBase Model=Qwen2.5-14B-Instruct2025.12 | 49.6 | |
| Qwen2.5-14B-InstructBase Model=Qwen2.5-14B-Instruct2025.12 | 48.5 | |
| Code Alpaca-1KBase Model=Qwen2.5-7B-Instruct2025.12 | 46.4 | |
| Self-OSSBase Model=Qwen2.5-7B-Instruct2025.12 | 45.9 | |
| DataFlow-Code-10KBase Model=Qwen2.5-7B-Instruct2025.12 | 45.4 | |
| DataFlow-Code-1KBase Model=Qwen2.5-7B-Instruct2025.12 | 45.1 | |
| DataFlow-Code-5KBase Model=Qwen2.5-7B-Instruct2025.12 | 45 | |
| Qwen2.5-7B-InstructBase Model=Qwen2.5-7B-Instruct2025.12 | 43.9 |