List your company

Multitask Language Understanding on MMLU (vbal)

65.7vbal ScoreBase
-2.62815.11132.8550.589Apr 6, 2026

Evaluation results

MethodLinks
BaseModel=Tulu3.1-8B2026.0465.7
BaseModel=OLMo2-7B2026.0464
BaseModel=Llama3.1-8B2026.0462.5
Oracle-rephraseModel=Llama3.1-8B2026.048.6
Oracle-rephraseModel=Tulu3.1-8B2026.046.6
Oracle-rephraseModel=OLMo2-7B2026.044.5
SIMPL.Model=Llama3.1-8B2026.044.3
STRUCT.Model=Llama3.1-8B2026.043.7
Reflect-and-rephraseModel=Llama3.1-8B2026.043.6
PROF.Model=Llama3.1-8B2026.042.9
PROF.Model=OLMo2-7B2026.041.9
STRUCT.Model=OLMo2-7B2026.041.3
PROF.Model=Tulu3.1-8B2026.040.9
Reflect-and-rephraseModel=OLMo2-7B2026.040.6
STRUCT.Model=Tulu3.1-8B2026.040.2
SIMPL.Model=OLMo2-7B2026.040.2
SIMPL.Model=Tulu3.1-8B2026.040
Reflect-and-rephraseModel=Tulu3.1-8B2026.040