Software Engineering on SWE tasks Configuration Integration n = 20 (test)
59.27F1 ScoreParametricSkills
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| ParametricSkillsBackbone=Qwen3-8B, Skill Type=Parametric (Hypernetwork generated)2026.06 | 59.27 | 90.19 | 77.75 | |
| In-ContextBackbone=Qwen3-8B, Skill Type=In-context skills2026.06 | 54.32 | 89.68 | 74 | |
| No-SkillBackbone=Qwen3-8B, Skill Type=None2026.06 | 53.87 | 89.32 | 66.75 | |
| SHINEBackbone=Qwen3-8B, Skill Type=Parametric (SHINE checkpoint)2026.06 | 43.31 | 88.97 | 61.25 |