Code Generation on MBPP-ET (Acc, RSR)
73AccEG-CFG
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| EG-CFGModel=DeepSeek-V3-03242025.06 | 73 | 23.29 | |
| MapCoderModel=DeepSeek-V3-03242025.06 | 69.6 | 13.63 | |
| LPWModel=GPT-4o2025.06 | 65.8 | — | |
| LPWModel=DeepSeek-V3-03242025.06 | 65.2 | 1.13 | |
| Baseline LLMModel=DeepSeek-V3-03242025.06 | 64.8 | 0 | |
| MGDebuggerModel=DeepSeek-V3-03242025.06 | 64.8 | 0 | |
| Self-CollaborationModel=GPT-42025.06 | 62.1 | — | |
| EG-CFGModel=DeepSeek-Coder 1.3B2025.06 | 59.8 | 29.96 | |
| MapCoderModel=GPT-42025.06 | 57.5 | — | |
| Baseline LLMModel=GPT-42025.06 | 49.2 | — | |
| MapCoderModel=DeepSeek-Coder 1.3B2025.06 | 46.2 | 6.27 | |
| MGDebuggerModel=DeepSeek-Coder 1.3B2025.06 | 44.6 | 3.48 | |
| Baseline LLMModel=DeepSeek-Coder 1.3B2025.06 | 42.6 | 0 |