Code Generation on Code Out-of-Domain
63.99AccuracyBase
Evaluation Results
| Method | Links | |
|---|---|---|
| BaseBase Model=Qwen2.5-1.5B2026.05 | 63.99 | |
| PionBase Model=Qwen2.5-1.5B2026.05 | 63.21 | |
| AdamWBase Model=Qwen2.5-1.5B2026.05 | 62.64 | |
| MuonBase Model=Qwen2.5-1.5B2026.05 | 62.41 | |
| PionBase Model=Llama3.2-3B2026.05 | 59.74 | |
| MuonBase Model=Llama3.2-3B2026.05 | 58.88 | |
| AdamWBase Model=Llama3.2-3B2026.05 | 58.64 | |
| BaseBase Model=Llama3.2-3B2026.05 | 53.08 |