List your company

Software Engineering on SWE tasks Diagnosis Search n = 30 (test)

53.68F1 ScoreParametricSkills
36.540840.990445.4449.8896Jun 29, 2026

Evaluation results

MethodLinks
ParametricSkillsBackbone=Qwen3-8B, Skill Type=Parametric (Hypernetwork generated)2026.0653.6889.3261.67
In-ContextBackbone=Qwen3-8B, Skill Type=In-context skills2026.0648.7588.3157.83
No-SkillBackbone=Qwen3-8B, Skill Type=None2026.0647.6188.3760.33
SHINEBackbone=Qwen3-8B, Skill Type=Parametric (SHINE checkpoint)2026.0637.287.5848.17