Question Answering on LiveMath
64.5AccuracySOFTSKILL
Evaluation Results
| Method | Links | |
|---|---|---|
| SOFTSKILLBase Model=Qwen3.5–4B, Insertion Placement=skill section2026.06 | 64.5 | |
| SOFTSKILL-meanBase Model=Qwen3.5–4B, Insertion Placement=prompt start2026.06 | 59.7 | |
| LoRABase Model=Qwen3.5–4B2026.06 | 58.9 | |
| SkillOptBase Model=Qwen3.5–4B2026.06 | 52 | |
| SOFTSKILLBase Model=Qwen3.5–4B, Insertion Placement=prompt start2026.06 | 50 | |
| SOFTSKILL-meanBase Model=Qwen3.5–4B, Insertion Placement=skill section2026.06 | 42.7 | |
| LLM skillBase Model=Qwen3.5–4B2026.06 | 28.8 | |
| GEPABase Model=Qwen3.5–4B2026.06 | 28.8 | |
| Trace2SkillBase Model=Qwen3.5–4B2026.06 | 27.2 | |
| No skillBase Model=Qwen3.5–4B2026.06 | 22.4 | |
| Human skillBase Model=Qwen3.5–4B2026.06 | 18.4 | |
| TextGradBase Model=Qwen3.5–4B2026.06 | 10.6 | |
| SOFTSKILL-mean-init-onlyBase Model=Qwen3.5–4B, Insertion Placement=prompt start2026.06 | 0.8 |