Clinical NLP Tasks on CASI CR (GPT-4 score)
3.71GPT-4 ScoreAsclepius-R
Evaluation Results
| Method | Links | |
|---|---|---|
| Asclepius-RModel Size=13B, Sample Size=1002023.09 | 3.71 | |
| Asclepius-RModel Size=7B, Sample Size=1002023.09 | 3.7 | |
| AsclepiusModel Size=13B, Sample Size=1002023.09 | 3.67 | |
| AsclepiusModel Size=7B, Sample Size=1002023.09 | 3.56 | |
| Clinical-CamelModel Size=13B, Sample Size=1002023.09 | 3.54 | |
| GPT-3.5-turboModel Size=175B, Sample Size=1002023.09 | 3.47 | |
| VicunaModel Size=13B, Sample Size=1002023.09 | 3.18 | |
| ChatDoctorModel Size=7B, Sample Size=1002023.09 | 2.56 | |
| AlpacaModel Size=7B, Sample Size=1002023.09 | 2.41 | |
| MedAlpacaModel Size=7B, Sample Size=1002023.09 | 2.16 |