Yes/No Medical Visual Question Answering on MIMIC
75.2F1 ScoreLobA
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| LobAModel Category=Adversarial Techniques2025.04 | 75.2 | 75.9 | 74.3 | |
| PH + LLaVA-MedModel Category=Adversarial Techniques2025.04 | 73.5 | 75.4 | 72.4 | |
| CRG + LLaVA-MedModel Category=Adversarial Techniques2025.04 | 73.1 | 71.5 | 72.5 | |
| LobA w/o self-promptModel Category=Adversarial Techniques, Self-prompt=false2025.04 | 72.7 | 71.6 | 73.1 | |
| LLaVA-MedModel Category=Open-source Models2025.04 | 72.2 | 71.2 | 70.9 | |
| LISAModel Category=Open-source Models2025.04 | 71.6 | 69.2 | 72.8 | |
| VCD + LLaVA-MedModel Category=Adversarial Techniques2025.04 | 71.6 | 70.8 | 72.1 | |
| BioMedGPTModel Category=Open-source Models2025.04 | 70.8 | 73.4 | 67.2 | |
| CheXagentModel Category=Open-source Models2025.04 | 67.7 | 62.1 | 69.5 | |
| GPT-4-OModel Category=Proprietary Models2025.04 | 64.6 | 70.6 | 55.6 | |
| MedFlamingoModel Category=Open-source Models2025.04 | 56.2 | 57.4 | 55.2 | |
| Gemini-1.5-Flash-8BModel Category=Proprietary Models2025.04 | 54.8 | 52 | 56.9 | |
| GPT-4-VisionModel Category=Proprietary Models2025.04 | 52.1 | 65.2 | 40.9 |