Question Answering on MedQA (dev)
77.6AccuracyKnowGPT
Evaluation results
| Method | Links | |
|---|---|---|
| KnowGPTCategory=Ours2023.12 | 77.6 | |
| MindmapCategory=LLM + KG Prompting2023.12 | 74.7 | |
| GPT-4Category=LLM + Zero-shot2023.12 | 73.9 | |
| RoGCategory=LLM + KG Prompting2023.12 | 71.3 | |
| CoKCategory=LLM + KG Prompting2023.12 | 70.6 | |
| Llama3Category=LLM + Zero-shot, Parameters=8b2023.12 | 63.9 | |
| GPT-3.5Category=LLM + Zero-shot, Implementation=gpt-3.5-turbo2023.12 | 48.4 | |
| ChatGLM2Category=LLM + Zero-shot2023.12 | 43.2 | |
| JointLKCategory=KG-enhanced LM2023.12 | 41.1 | |
| GrapeQACategory=KG-enhanced LM2023.12 | 40.1 | |
| GreaseLMCategory=KG-enhanced LM2023.12 | 40 | |
| HamQACategory=KG-enhanced LM2023.12 | 39.6 | |
| QA-GNNCategory=KG-enhanced LM2023.12 | 39.4 | |
| Bert-largeCategory=LM + Fine tuning2023.12 | 37.3 | |
| RoBerta-largeCategory=LM + Fine tuning2023.12 | 36.9 | |
| Bert-baseCategory=LM + Fine tuning2023.12 | 35.9 | |
| ChatGLMCategory=LLM + Zero-shot2023.12 | 34.6 | |
| Llama2Category=LLM + Zero-shot, Parameters=7b2023.12 | 33.8 | |
| BaichuanCategory=LLM + Zero-shot, Parameters=7B2023.12 | 33.4 | |
| InternLMCategory=LLM + Zero-shot2023.12 | 32.5 | |
| GPT-3Category=LLM + Zero-shot, Implementation=text-davinci-0022023.12 | 31.2 |