Medical Question Answering on MedXpertQA (test)
8.49ETS ScoreGPT4o
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| GPT4ocategory=Base Model2025.12 | 8.49 | 8.26 | 8.34 | 8.36 | |
| o3category=Base Model2025.12 | 8.37 | 8.18 | 8.52 | 8.42 | |
| MedTutor-R1category=Final Model2025.12 | 8.33 | 8.41 | 8.26 | 8.35 | |
| MedTutor-R1simulation_variation=w/o Specialist2025.12 | 8.32 | 8.29 | 7.81 | 8.14 | |
| MedTutor-R1training_variation=w/o reward veto2025.12 | 8.3 | 8.39 | 6.92 | 7.87 | |
| MedTutor-R1training_variation=w/o CS reward2025.12 | 8.28 | 8.38 | 7.55 | 8.07 | |
| MedTutor-R1simulation_variation=w/o Supervisor2025.12 | 8.19 | 8.43 | 7.73 | 8.08 | |
| MedTutor-R1training_variation=w/ LLaVA-based2025.12 | 8.13 | 8.16 | 7.85 | 8.05 | |
| DeepSeek-R1category=Base Model2025.12 | 8.12 | 7.84 | 8.07 | 8.01 | |
| MedTutor-R1training_variation=w/o AQ reward2025.12 | 8.12 | 7.45 | 8.22 | 7.93 | |
| MedTutor-R1training_variation=w/ Vanilla reward2025.12 | 8.05 | 7.9 | 8.07 | 8.01 | |
| MedTutor-R1training_variation=w/o IS reward2025.12 | 7.95 | 8.32 | 8.2 | 8.16 | |
| MedTutor-R1simulation_variation=w/o Patient2025.12 | 7.91 | 8.16 | 7.88 | 7.98 | |
| MedTutor-R1training_variation=w/o Thinking2025.12 | 7.82 | 8.06 | 7.93 | 7.94 | |
| MedTutor-R1training_variation=w/o RL2025.12 | 7.58 | 7.83 | 7.65 | 7.69 | |
| MedTutor-R1simulation_variation=w One-Student2025.12 | 7.58 | 7.69 | 8.21 | 7.86 | |
| TutorRLcategory=Agent for Medical Education2025.12 | 7.5 | 7.49 | 7.26 | 7.42 | |
| Med-SocraticLMcategory=Agent for Medical Education2025.12 | 7.26 | 7.33 | 7.64 | 7.41 | |
| Qwen2.5VLcategory=Base Model2025.12 | 7.07 | 7.04 | 6.78 | 6.96 | |
| DRLTutorcategory=Agent for Medical Education2025.12 | 6.98 | 7.58 | 7.43 | 7.32 | |
| EduChat-R1category=Agent for Medical Education2025.12 | 6.88 | 6.95 | 7.37 | 7.07 | |
| InternVL-3.5category=Base Model2025.12 | 6.82 | 6.65 | 6.83 | 6.77 | |
| LLaVA-V1.6category=Base Model2025.12 | 5.45 | 6.15 | 6.02 | 5.87 |