Mathematical Reasoning on GSM8K (vbal)
0.773vbal ScoreBase
Evaluation Results
| Method | Links | |
|---|---|---|
| BaseModel=Llama3.1-8B2026.04 | 0.773 | |
| BaseModel=Tulu3.1-8B2026.04 | 0.757 | |
| BaseModel=OLMo2-7B2026.04 | 0.736 | |
| Reflect-and-rephraseModel=OLMo2-7B2026.04 | 0.083 | |
| Reflect-and-rephraseModel=Tulu3.1-8B2026.04 | 0.048 | |
| Reflect-and-rephraseModel=Llama3.1-8B2026.04 | 0.027 | |
| PROF.Model=OLMo2-7B2026.04 | -0.015 | |
| STRUCT.Model=OLMo2-7B2026.04 | -0.017 | |
| PROF.Model=Tulu3.1-8B2026.04 | -0.023 | |
| STRUCT.Model=Tulu3.1-8B2026.04 | -0.03 | |
| PROF.Model=Llama3.1-8B2026.04 | -0.04 | |
| STRUCT.Model=Llama3.1-8B2026.04 | -0.047 | |
| SIMPL.Model=OLMo2-7B2026.04 | -0.058 | |
| SIMPL.Model=Tulu3.1-8B2026.04 | -0.059 | |
| SIMPL.Model=Llama3.1-8B2026.04 | -0.08 |