Medical Question Answering on Huatuo-26M-Lite-100
83.25Win RateLLM-AutoDP
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| LLM-AutoDPBackbone=Llama3.1-8B, Baseline Comparison=No-Process2026.01 | 83.25 | 0.25 | 16.5 | |
| LLM-AutoDPBackbone=Qwen2.5-7B, Baseline Comparison=No-Process2026.01 | 83.12 | 2.5 | 14.38 | |
| LLM-AutoDPBackbone=Gemma-2-9B, Baseline Comparison=No-Process2026.01 | 82.37 | 0 | 17.63 | |
| LLM-AutoDPBackbone=Gemma-2-9B, Baseline Comparison=RS2026.01 | 61.37 | 2.75 | 35.88 | |
| LLM-AutoDPBackbone=Qwen2.5-7B, Baseline Comparison=SELA2026.01 | 61.25 | 8.25 | 30.25 | |
| LLM-AutoDPBackbone=Llama3.1-8B, Baseline Comparison=RS2026.01 | 60 | 2.5 | 37.25 | |
| LLM-AutoDPBackbone=Qwen2.5-7B, Baseline Comparison=RS2026.01 | 59.5 | 11 | 29.5 | |
| LLM-AutoDPBackbone=Gemma-2-9B, Baseline Comparison=All-Process2026.01 | 57.73 | 2.25 | 40.02 | |
| LLM-AutoDPBackbone=Gemma-2-9B, Baseline Comparison=SELA2026.01 | 57.66 | 4.25 | 38.09 | |
| LLM-AutoDPBackbone=Llama3.1-8B, Baseline Comparison=All-Process2026.01 | 57.5 | 2.75 | 39.75 | |
| LLM-AutoDPBackbone=Qwen2.5-7B, Baseline Comparison=All-Process2026.01 | 55.75 | 11.5 | 32.75 | |
| LLM-AutoDPBackbone=Llama3.1-8B, Baseline Comparison=SELA2026.01 | 54.75 | 2.75 | 42.5 |