List your company
ResearchFormerly WizWand

Clinical NLP Tasks on i2b2 (GPT-4 score)

3.36GPT-4 ScoreGPT-3.5-turbo
1.30081.83542.372.9046Sep 1, 2023
Updated 8mo ago

Evaluation results

MethodLinks
GPT-3.5-turboModel Size=175B, Sample Size=6192023.093.36
Asclepius-RModel Size=13B, Sample Size=6192023.093.35
AsclepiusModel Size=13B, Sample Size=6192023.093.34
Asclepius-RModel Size=7B, Sample Size=6192023.093.32
AsclepiusModel Size=7B, Sample Size=6192023.093.25
Clinical-CamelModel Size=13B, Sample Size=6192023.093.25
VicunaModel Size=13B, Sample Size=6192023.093.08
AlpacaModel Size=7B, Sample Size=6192023.092.47
ChatDoctorModel Size=7B, Sample Size=6192023.092.43
MedAlpacaModel Size=7B, Sample Size=6192023.091.38