Long-form QA on MedicationQA (test)
27.73ROUGE-1Fine-Tuned GPT-4o + MedBioRAG
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Fine-Tuned GPT-4o + MedBioRAGFine-tuned=true, MedBioRAG=true2025.12 | 27.73 | 15.09 | 22.72 | 7.24 | 8.79 | -33.63 | |
| Fine-Tuned GPT-4oFine-tuned=true, MedBioRAG=false2025.12 | 24.69 | 8.8 | 17.61 | 2.49 | 8.98 | -33.82 | |
| GPT-4oFine-tuned=false, MedBioRAG=false2025.12 | 22.92 | 13.69 | 18.7 | 7.89 | 8.55 | -6.92 | |
| GPT-4o + MedBioRAGFine-tuned=false, MedBioRAG=true2025.12 | 19.85 | 4.2 | 10.97 | 0.98 | -7.63 | -33.21 |