Disease Diagnosis on CMCQA
37.18Win RateMindMap
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| MindMapComparison Baseline=Embedding Retriever, Evaluator=GPT-42023.08 | 37.18 | 42.74 | 20.08 | |
| MindMapComparison Baseline=GPT-3.5, Evaluator=GPT-42023.08 | 35.68 | 39.96 | 24.36 | |
| MindMapComparison Baseline=KG Retriever, Evaluator=GPT-42023.08 | 34.4 | 45.51 | 20.09 | |
| MindMapComparison Baseline=BM25 Retriever, Evaluator=GPT-42023.08 | 30.98 | 42.74 | 26.28 | |
| MindMapComparison Baseline=GPT-4, Evaluator=GPT-42023.08 | 27.99 | 47.22 | 24.79 |