List your company

Out-of-Distribution Detection on OBQA to MMLU

87.09AUROCETN
44.741255.735666.7377.7244Apr 9, 2026

Evaluation results

MethodLinks
ETNBackbone=Llama-3.1-8B, Measure=MP2026.0487.09—
ETNBackbone=Llama-3.1-8B, Measure=DE2026.0486.27—
ETNBackbone=Gemma-2-9B, Measure=MP2026.0485.67—
ETNBackbone=Llama-3.1-8B, Measure=MI2026.0485.3—
MAP_CEBackbone=Llama-3.1-8B, Measure=MP2026.0483.48—
MAP_CEBackbone=Gemma-2-9B, Measure=MP2026.0483.48—
ETNscoring=maximum probability2026.0483.3991.7
IB-EDLBackbone=Llama-3.1-8B, Measure=DE2026.0483.16—
IB-EDLBackbone=Llama-3.1-8B, Measure=MP2026.0482.83—
MCDBackbone=Llama-3.1-8B, Measure=MP2026.0481.96—
ETNBackbone=Gemma-2-9B, Measure=DE2026.0480.92—
DeepEnsBackbone=Llama-3.1-8B, Measure=MP2026.0480.33—
MCDBackbone=Gemma-2-9B, Measure=MP2026.0478.99—
DMMBackbone=Llama-3.1-8B, Measure=MP2026.0478.98—
DMMBackbone=Gemma-2-9B, Measure=MP2026.0478.98—
ETNBackbone=Gemma-2-9B, Measure=MI2026.0478.82—
DMMBackbone=Llama-3.1-8B, Measure=DE2026.0478.47—
DMMBackbone=Gemma-2-9B, Measure=DE2026.0478.47—
DeepEnsBackbone=Llama-3.1-8B, Measure=MI2026.0475.67—
DeepEnsBackbone=Gemma-2-9B, Measure=MP2026.0475.44—
MCDBackbone=Gemma-2-9B, Measure=MI2026.0475.31—
IB-EDLBackbone=Gemma-2-9B, Measure=MP2026.0473.78—
DMMBackbone=Llama-3.1-8B, Measure=MI2026.0473.51—
DMMBackbone=Gemma-2-9B, Measure=MI2026.0473.51—
MAP_EDLBackbone=Gemma-2-9B, Measure=DE2026.0473.41—
MAP_EDLBackbone=Gemma-2-9B, Measure=MP2026.0473.38—
MAP_EDLBackbone=Llama-3.1-8B, Measure=DE2026.0472.53—
MAP_EDLBackbone=Llama-3.1-8B, Measure=MP2026.0472.17—
MCDBackbone=Llama-3.1-8B, Measure=MI2026.0469.58—
IB-EDLBackbone=Gemma-2-9B, Measure=DE2026.0468.66—
IB-EDLBackbone=Llama-3.1-8B, Measure=MI2026.0467.89—
DeepEnsBackbone=Gemma-2-9B, Measure=MI2026.0467.04—
MAP_EDLBackbone=Llama-3.1-8B, Measure=MI2026.0466.99—
IB-EDLBackbone=Gemma-2-9B, Measure=MI2026.0461.76—
MAP_EDLBackbone=Gemma-2-9B, Measure=MI2026.0459.13—
MD2026.0454.4270.5
LLBackbone=Gemma-2-9B, Measure=MI2026.0451.54—
ODIN2026.0450.1161.35
LLBackbone=Llama-3.1-8B, Measure=MP2026.0447.43—
LLBackbone=Gemma-2-9B, Measure=MP2026.0447.43—
LLBackbone=Llama-3.1-8B, Measure=MI2026.0446.37—