Clinical NLP Tasks on MIMIC-III (GPT-4 Score)
3.48GPT-4 ScoreAsclepius-R
Evaluation Results
| Method | Links | |
|---|---|---|
| Asclepius-RModel Size=13B, Sample Size=4132023.09 | 3.48 | |
| GPT-3.5-turboModel Size=175B, Sample Size=4132023.09 | 3.46 | |
| Asclepius-RModel Size=7B, Sample Size=4132023.09 | 3.43 | |
| AsclepiusModel Size=13B, Sample Size=4132023.09 | 3.36 | |
| AsclepiusModel Size=7B, Sample Size=4132023.09 | 3.21 | |
| Clinical-CamelModel Size=13B, Sample Size=4132023.09 | 2.93 | |
| VicunaModel Size=13B, Sample Size=4132023.09 | 2.9 | |
| ChatDoctorModel Size=7B, Sample Size=4132023.09 | 2.21 | |
| AlpacaModel Size=7B, Sample Size=4132023.09 | 1.95 | |
| MedAlpacaModel Size=7B, Sample Size=4132023.09 | 1.45 |