Arithmetic function acquisition on KGFUNC (test)
90.9AccuracyOPSD
Evaluation Results
| Method | Links | |
|---|---|---|
| OPSDModel=Qwen3-4B-It2026.05 | 90.9 | |
| OPSD-NLLModel=Qwen3-4B-It, n=1, T=02026.05 | 89.1 | |
| MIXSDModel=Qwen3-4B-It, lambda=0.32026.05 | 89.1 | |
| MIXSDModel=Qwen3-8B, lambda=0.32026.05 | 89.1 | |
| SFTModel=Qwen3-8B2026.05 | 86.9 | |
| OPSD-NLLModel=Qwen3-8B, n=1, T=02026.05 | 86.9 | |
| OPSDModel=Qwen3-8B2026.05 | 80 | |
| MIXSDModel=Qwen3-8B, lambda=02026.05 | 78.3 | |
| MIXSDModel=Qwen3-4B-It, lambda=02026.05 | 77.1 | |
| SFTModel=Qwen3-4B-It2026.05 | 72.6 | |
| MIXSDModel=Qwen3-4B-It, lambda=0.52026.05 | 56.6 | |
| OPSDModel=Qwen3-1.7B2026.05 | 54.3 | |
| MIXSDModel=Qwen3-8B, lambda=0.52026.05 | 52 | |
| SFTModel=Qwen3-1.7B2026.05 | 51.4 | |
| MIXSDModel=Qwen3-1.7B, lambda=0.32026.05 | 45.7 | |
| MIXSDModel=Qwen3-1.7B, lambda=02026.05 | 44 | |
| OPSD-NLLModel=Qwen3-1.7B, n=1, T=02026.05 | 25.7 | |
| MIXSDModel=Qwen3-1.7B, lambda=0.52026.05 | 18.3 | |
| BaseModel=Qwen3-8B2026.05 | 2.9 | |
| BaseModel=Qwen3-1.7B2026.05 | 1.7 | |
| BaseModel=Qwen3-4B-It2026.05 | 0 |