Code Generation on Dafny2Verus
81.46P@1ANCORA
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ANCORABackbone=Qwen2.5-Coder-3B-Instruct, Evaluation Setting=Test-Time Training, Seed Configuration=Test Dataset, Prompting=0-shot2026.04 | 81.46 | 83.86 | 84.58 | |
| PSVBackbone=Qwen2.5-Coder-3B-Instruct, Evaluation Setting=Test-Time Training, Seed Configuration=Test Dataset, Prompting=1-shot2026.04 | 65.63 | 78.04 | 80.06 | |
| RFTBackbone=Qwen2.5-Coder-3B-Instruct, Evaluation Setting=Test-Time Training, Seed Configuration=Test Dataset, Prompting=1-shot2026.04 | 34.46 | 57.12 | 63.4 | |
| AlphaVerusBackbone=Qwen2.5-Coder-3B-Instruct, Evaluation Setting=Transfer Learning, Seed Configuration=Dafny2Verus, Prompting=50-shot2026.04 | 24.06 | 52.42 | 63.44 | |
| AlphaVerusBackbone=Qwen2.5-Coder-3B-Instruct, Evaluation Setting=Test-Time Training, Seed Configuration=Test Dataset, Prompting=50-shot2026.04 | 24.06 | 52.42 | 63.44 |