Software Engineering Skill Evaluation on SWE-Skills-Bench risk-metrics-calculation
5Pass RateVanilla LLM
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Vanilla LLMSetting=Vanilla baseline2026.06 | 5 | — | — | |
| S2LSetting=Our method2026.06 | 5 | 0.0912 | 0 | |
| Full Skill TextSetting=Full Skill Text prompting2026.06 | 1 | 0.721 | -0.55 |