Medical Dialogue Generation on KaMed
41.88BLEU-1GPT-4o
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| GPT-4oCategory=LLM-based2025.06 | 41.88 | — | 23.34 | 13.86 | 13.94 | 3.1 | |
| MedRefCategory=LLM-based, Backbone=ChatGLM3-6B, Fine-tuning=LoRA2025.06 | 40.47 | 31.62 | 21.28 | 21.96 | 28.14 | 12.42 | |
| HuatuoGPT-IICategory=LLM-based, Backbone=Baichuan-7B2025.06 | 40.35 | 32.93 | 23.92 | 12 | 13.84 | 2.74 | |
| DFMedCategory=PLM-based2025.06 | 39.59 | 30.53 | 20.3 | 21.33 | 27.67 | 11.21 | |
| DISC-MedLLMCategory=LLM-based, Backbone=Baichuan-13B2025.06 | 38.05 | — | 20.26 | 13.54 | 20.48 | 5.93 | |
| GPT-2Category=PLM-based2025.06 | 33.76 | 26.58 | 17.82 | 17.26 | 26.8 | 10.56 | |
| BARTCategory=PLM-based2025.06 | 33.62 | 26.43 | 17.64 | 19.2 | 27.91 | 11.43 | |
| Chatglm3-6BCategory=LLM-based2025.06 | 32.03 | 25.2 | 16.68 | 20.56 | 28.02 | 12.12 | |
| VRBotCategory=DL-based2025.06 | 30.04 | 23.76 | 16.36 | 12.08 | 18.71 | 7.28 | |
| ZhongjingCategory=LLM-based2025.06 | 27.48 | 22.35 | 15.52 | 6.44 | 13.7 | 3.05 | |
| Seq2SeqCategory=DL-based2025.06 | 23.52 | 18.56 | 12.13 | — | 23.56 | 8.67 |