List your company

Medical Question Answering on PubMedQA (FA, HS)

95.63Factual Accuracy (FA)SPPFT
27.655645.302862.9580.5972Apr 14, 2026

Evaluation results

MethodLinks
SPPFTModel=Llama-3-8B2026.0495.6325.45
CWACModel=Llama-3-8B2026.0495.528.65
SFTModel=Llama-3-8B2026.0495.2453.1
CWACModel=Gemma-2-9B2026.0495.212.55
ASFTModel=Llama-3-8B2026.0495.1810.28
CWACModel=Llama-2-7B2026.0494.527.24
ASFTModel=Gemma-2-9B2026.0494.4111.63
BEAModel=Llama-2-7B2026.0494.335.17
BEAModel=Llama-3-8B2026.0494.2729.12
BEAModel=Gemma-2-9B2026.0494.1825.31
SFTModel=Llama-2-7B2026.0493.8146.75
SFTModel=Gemma-2-9B2026.0493.4548.97
SPPFTModel=Gemma-2-9B2026.0493.2616.49
ASFTModel=Llama-2-7B2026.0493.229.38
LisaModel=Gemma-2-9B2026.0492.0527.4
CWACModel=Mistral-7B2026.0490.7215.39
LisaModel=Llama-2-7B2026.0490.6222.37
LisaModel=Llama-3-8B2026.0487.630.41
SPPFTModel=Llama-2-7B2026.0485.3716.42
SafeInstrModel=Gemma-2-9B2026.0483.459.54
SafeInstrModel=Llama-3-8B2026.0474.0515.37
SafeInstrModel=Llama-2-7B2026.0470.3712.49
SFTModel=Mistral-7B2026.0464.5360.72
BEAModel=Mistral-7B2026.0462.6745.25
ASFTModel=Mistral-7B2026.0458.1520.31
LisaModel=Mistral-7B2026.0443.543.41
SafeInstrModel=Mistral-7B2026.0443.0845.37
SPPFTModel=Mistral-7B2026.0430.2729.46