Code Generation on APPS (tau and Speedup evaluation)
5.65TauEDA (Ours)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| EDA (Ours)Temperature=0, Target Model=Qwen2.5-Coder-7B2026.03 | 5.65 | 3.34 | |
| EDA (Base)Temperature=0, Target Model=Qwen2.5-Coder-7B2026.03 | 4.92 | 2.9 | |
| Full-FTTemperature=0, Target Model=Qwen2.5-Coder-7B2026.03 | 4.87 | 2.91 | |
| LoRATemperature=0, Target Model=Qwen2.5-Coder-7B2026.03 | 4.8 | 2.88 | |
| EDA (Ours)Temperature=1, Target Model=Qwen2.5-Coder-7B2026.03 | 4.62 | 2.38 | |
| Full-FTTemperature=1, Target Model=Qwen2.5-Coder-7B2026.03 | 4.05 | 2.13 | |
| EDA (Base)Temperature=1, Target Model=Qwen2.5-Coder-7B2026.03 | 4 | 2.1 | |
| LoRATemperature=1, Target Model=Qwen2.5-Coder-7B2026.03 | 3.95 | 2.08 | |
| Training-FreeTemperature=0, Target Model=Qwen2.5-Coder-7B2026.03 | 1.69 | 1.14 | |
| Training-FreeTemperature=1, Target Model=Qwen2.5-Coder-7B2026.03 | 1.35 | 0.81 |