Software Engineering on SWE tasks Feature Patch n = 125 (test)
55.72F1 ScoreParametricSkills
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ParametricSkillsBackbone=Qwen3-8B, Skill Type=Parametric (Hypernetwork generated)2026.06 | 55.72 | 89.83 | 74.2 | |
| In-ContextBackbone=Qwen3-8B, Skill Type=In-context skills2026.06 | 49.86 | 88.51 | 68.12 | |
| No-SkillBackbone=Qwen3-8B, Skill Type=None2026.06 | 48.9 | 88.92 | 66.68 | |
| SHINEBackbone=Qwen3-8B, Skill Type=Parametric (SHINE checkpoint)2026.06 | 39.41 | 88.15 | 56.36 |