Statements (23)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:large_language_model
|
| gptkbp:accuracyOnUSMLE |
67.6%
|
| gptkbp:announced |
2022
|
| gptkbp:basedOn |
gptkb:PaLM
|
| gptkbp:contrastsWith |
gptkb:ChatGPT
|
| gptkbp:developedBy |
gptkb:Google_Research
|
| gptkbp:domain |
medical
|
| gptkbp:evaluatedOn |
gptkb:USMLE
|
| gptkbp:focusesOn |
medical reasoning
medical factuality medical safety |
| gptkbp:language |
English
|
| gptkbp:notablePublication |
gptkb:Large_Language_Models_Encode_Clinical_Knowledge
|
| gptkbp:outperformed |
gptkb:GPT-3
|
| gptkbp:paperPublishedYear |
2023
|
| gptkbp:publishedIn |
gptkb:Nature
|
| gptkbp:purpose |
medical question answering
|
| gptkbp:successor |
gptkb:Med-PaLM_2
|
| gptkbp:uses |
instruction tuning
medical expert feedback |
| gptkbp:bfsParent |
gptkb:Google_AI
|
| gptkbp:bfsLayer |
6
|
| http://www.w3.org/2000/01/rdf-schema#label |
Med-PaLM
|