Physician Evaluation of Long-Form Answers on Multi-MedQA 140
92.1Consensus SupportPhysician
Evaluation results
| Method | Links | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| PhysicianEvaluation=Physician rater evaluation2023.05 | 92.1 | 92.9 | 97.1 | 97.1 | 97.1 | 97.1 | 97.1 | 95 | 96.4 | 97.1 | 87.1 | 97.1 | — | — | — | |
| Med-PaLM 2Evaluation=Physician rater evaluation2023.05 | 91.7 | 93.3 | 95.5 | 98.3 | 97.1 | 97.4 | 98.6 | 93.3 | 96.2 | 90 | 88.1 | 97.1 | — | — | — | |
| Med-PaLMEvaluation=Physician rater evaluation2023.05 | 82.9 | 94.3 | 97.9 | 93.6 | 93.6 | 91.4 | 94.3 | 82.9 | 88.6 | 81.4 | 85 | 99.3 | — | — | — |