List your company

Mathematical Reasoning on GSM8K (FA, HS)

89.47FACWAC
26.997243.216159.43575.6539Apr 14, 2026

Evaluation results

MethodLinks
CWACModel=Gemma-2-9B2026.0489.478.38
LisaModel=Gemma-2-9B2026.0488.2543.48
SFTModel=Gemma-2-9B2026.0487.655.42
ASFTModel=Gemma-2-9B2026.0483.3147.53
CWACModel=Llama-3-8B2026.0478.137.19
BEAModel=Gemma-2-9B2026.0475.9654.21
SFTModel=Llama-3-8B2026.0475.8966.74
SPPFTModel=Gemma-2-9B2026.0473.225.64
CWACModel=Llama-2-7B2026.0472.5612.54
SFTModel=Mistral-7B2026.0472.0679.82
SFTModel=Llama-2-7B2026.0470.8152.17
CWACModel=Mistral-7B2026.0468.7625.63
BEAModel=Mistral-7B2026.0464.2753.32
SafeInstrModel=Gemma-2-9B2026.0462.7338.92
ASFTModel=Llama-2-7B2026.0457.3620.31
SPPFTModel=Llama-3-8B2026.0454.6715.27
BEAModel=Llama-2-7B2026.0445.5438.57
BEAModel=Llama-3-8B2026.0443.5732.24
SafeInstrModel=Mistral-7B2026.0443.2532.58
ASFTModel=Llama-3-8B2026.0440.3523.78
ASFTModel=Mistral-7B2026.0439.7645.63
LisaModel=Mistral-7B2026.0438.6256.7
SPPFTModel=Llama-2-7B2026.0438.3225.58
SPPFTModel=Mistral-7B2026.0437.8354.27
LisaModel=Llama-3-8B2026.0435.3428.42
LisaModel=Llama-2-7B2026.0434.2726.34
SafeInstrModel=Llama-2-7B2026.0432.2917.68
SafeInstrModel=Llama-3-8B2026.0429.420.15